Systematic review of the clinical effectiveness and cost-effectiveness ...

Health Technology Assessment 2010; Vol. 14: No. 4Systematic review of the clinicaleffectiveness and cost-effectivenessof photodynamic diagnosis and urinebiomarkers (FISH, ImmunoCyt, NMP22)and cytology for the detection andfollow-up of bladder cancerG Mowatt, S Zhu, M Kilonzo,C Boachie, C Fraser, TRL Griffiths,J N’Dow, G Nabi, J Cook and L ValeJanuary 2010DOI: 10.3310/hta14040Health Technology AssessmentNIHR HTA programmewww.hta.ac.uk

HTAHow to obtain copies of this and other HTA programme reportsAn electronic version of this title, in Adobe Acrobat format, is available for downloading free of chargefor personal use from the HTA website (www.hta.ac.uk). A fully searchable DVD is also available (seebelow).Printed copies of HTA journal series issues cost £20 each (post and packing free in the UK) to bothpublic and private sector purchasers from our despatch agents.Non-UK purchasers will have to pay a small fee for post and packing. For European countries the costis £2 per issue and for the rest of the world £3 per issue.How to order:– fax (with credit card details)– post (with credit card details or cheque)– phone during office hours (credit card only).Additionally the HTA website allows you to either print out your order or download a blank orderform.Contact details are as follows:Synergie UK (HTA Department)Digital House, The Loddon CentreWade RoadBasingstokeHants RG24 8QWEmail: orders@hta.ac.ukTel: 0845 812 4000 – ask for ‘HTA Payment Services’(out-of-hours answer-phone service)Fax: 0845 812 4001 – put ‘HTA Order’ on the fax headerPayment methodsPaying by chequeIf you pay by cheque, the cheque must be in pounds sterling, made payable to University ofSouthampton and drawn on a bank with a UK address.Paying by credit cardYou can order using your credit card by phone, fax or post.SubscriptionsNHS libraries can subscribe free of charge. Public libraries can subscribe at a reduced cost of £100 foreach volume (normally comprising 40–50 titles). The commercial subscription rate is £400 per volume(addresses within the UK) and £600 per volume (addresses outside the UK). Please see our website fordetails. Subscriptions can be purchased only for the current or forthcoming volume.How do I get a copy of HTA on DVD?Please use the form on the HTA website (www.hta.ac.uk/htacd/index.shtml). HTA on DVD is currentlyfree of charge worldwide.The website also provides information about the HTA programme and lists the membership of the variouscommittees.

NIHR Health Technology Assessment programmeThe Health Technology Assessment (HTA) programme, part of the National Institute for HealthResearch (NIHR), was set up in 1993. It produces high-quality research information on theeffectiveness, costs and broader impact of health technologies for those who use, manage and provide carein the NHS. ‘Health technologies’ are broadly defined as all interventions used to promote health, preventand treat disease, and improve rehabilitation and long-term care.The research findings from the HTA programme directly influence decision-making bodies such as theNational Institute for Health and Clinical Excellence (NICE) and the National Screening Committee(NSC). HTA findings also help to improve the quality of clinical practice in the NHS indirectly in that theyform a key component of the ‘National Knowledge Service’.The HTA programme is needs led in that it fills gaps in the evidence needed by the NHS. There are threeroutes to the start of projects.First is the commissioned route. Suggestions for research are actively sought from people working in theNHS, from the public and consumer groups and from professional bodies such as royal colleges and NHStrusts. These suggestions are carefully prioritised by panels of independent experts (including NHS serviceusers). The HTA programme then commissions the research by competitive tender.Second, the HTA programme provides grants for clinical trials for researchers who identify researchquestions. These are assessed for importance to patients and the NHS, and scientific rigour.Third, through its Technology Assessment Report (TAR) call-off contract, the HTA programmecommissions bespoke reports, principally for NICE, but also for other policy-makers. TARs bring togetherevidence on the value of specific technologies.Some HTA research projects, including TARs, may take only months, others need several years. Theycan cost from as little as £40,000 to over £1 million, and may involve synthesising existing evidence,undertaking a trial, or other research collecting new data to answer a research problem.The final reports from HTA projects are peer reviewed by a number of independent expert referees beforepublication in the widely read journal series Health Technology Assessment.Criteria for inclusion in the HTA journal seriesReports are published in the HTA journal series if (1) they have resulted from work for the HTAprogramme, and (2) they are of a sufficiently high scientific quality as assessed by the referees andeditors.Reviews in Health Technology Assessment are termed ‘systematic’ when the account of the search, appraisaland synthesis methods (to minimise biases and random errors) would, in theory, permit the replicationof the review by others.The research reported in this issue of the journal was commissioned and funded by the HTA programmeon behalf of NICE as project number 07/02/01. The protocol was agreed in October 2007. The assessmentreport began editorial review in October 2008 and was accepted for publication in May 2009. The authorshave been wholly responsible for all data collection, analysis and interpretation, and for writing up theirwork. The HTA editors and publisher have tried to ensure the accuracy of the authors’ report and wouldlike to thank the referees for their constructive comments on the draft document. However, they do notaccept liability for damages or losses arising from material published in this report.The views expressed in this publication are those of the authors and not necessarily those of the HTAprogramme or the Department of Health.Editor-in-Chief:Series Editors:Professor Tom Walley CBEDr Martin Ashton-Key, Dr Aileen Clarke, Professor Chris Hyde,Dr Tom Marshall, Dr John Powell, Dr Rob Riemsma and Professor Ken SteinISSN 1366-5278© 2010 Queen’s Printer and Controller of HMSOThis journal may be freely reproduced for the purposes of private research and study and may be included in professional journals provided thatsuitable acknowledgement is made and the reproduction is not associated with any form of advertising.Applications for commercial reproduction should be addressed to: NETSCC, Health Technology Assessment, Alpha House, University ofSouthampton Science Park, Southampton SO16 7NS, UK.Published by Prepress Projects Ltd, Perth, Scotland (www.prepress-projects.co.uk), on behalf of NETSCC, HTA.Printed on acid-free paper in the UK by Henry Ling Ltd, The Dorset Press, Dorchester.T

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4AbstractSystematic review of the clinical effectiveness andcost-effectiveness of photodynamic diagnosis and urinebiomarkers (FISH, ImmunoCyt, NMP22) and cytologyfor the detection and follow-up of bladder cancerG Mowatt, 1 * S Zhu, 1 M Kilonzo, 2 C Boachie, 1 C Fraser, 1 TRL Griffiths, 3J N’Dow, 4 G Nabi, 4 J Cook 1 and L Vale 1,21Health Services Research Unit, Institute of Applied Health Sciences, University of Aberdeen, UK2Health Economics Research Unit, Institute of Applied Health Sciences, University of Aberdeen, UK3Department of Cancer Studies and Molecular Medicine, University of Leicester, UK4Academic Urology Unit, Department of Surgery, University of Aberdeen, UK*Corresponding authorObjective: To assess the clinical effectiveness andcost-effectiveness of photodynamic diagnosis (PDD)compared with white light cystoscopy (WLC), and urinebiomarkers [fluorescence in situ hybridisation (FISH),ImmunoCyt, NMP22] and cytology for the detectionand follow-up of bladder cancer.Data sources: Major electronic databases includingMEDLINE, MEDLINE In-Process, EMBASE, BIOSIS,Science Citation Index, Health Management InformationConsortium and the Cochrane Controlled TrialsRegister were searched until April 2008.Review methods: A systematic review of theliterature was carried out according to standardmethods. An economic model was constructed toassess the cost-effectiveness of alternative diagnosticand follow-up strategies for the diagnosis andmanagement of patients with bladder cancer.Results: In total, 27 studies reported PDD testperformance. In pooled estimates [95% confidenceinterval (CI)] for patient-level analysis, PDD had highersensitivity than WLC [92% (80% to 100%) versus 71%(49% to 93%)] but lower specificity [57% (36% to79%) versus 72% (47% to 96%)]. Similar results werefound for biopsy-level analysis. The median sensitivities(range) of PDD and WLC for detecting lower risk,less aggressive tumours were similar for patient-leveldetection [92% (20% to 95%) versus 95% (8% to 100%)],but sensitivity was higher for PDD than for WLC forbiopsy-level detection [96% (88% to 100%) versus88% (74% to 100%)]. For more aggressive, higher-risktumours the median sensitivity of PDD for both patientlevel[89% (6% to 100%)] and biopsy-level [99% (54% to100%)] detection was higher than those of WLC [56%(0% to 100%) and 67% (0% to 100%) respectively]. FourRCTs comparing PDD with WLC reported effectivenessoutcomes. PDD use at transurethral resection ofbladder tumour resulted in fewer residual tumours atcheck cystoscopy [relative risk, RR, 0.37 (95% CI 0.20to 0.69)] and longer recurrence-free survival [RR 1.37(95% CI 1.18 to 1.59)] compared with WLC. In 71studies reporting the performance of biomarkers andcytology in detecting bladder cancer, sensitivity (95%CI) was highest for ImmunoCyt [84% (77% to 91%)]and lowest for cytology [44% (38% to 51%)], whereasspecificity was highest for cytology [96% (94% to 98%)]and lowest for ImmunoCyt [75% (68% to 83%)]. In thecost-effectiveness analysis the most effective strategy interms of true positive cases (44) and life-years (11.66)[flexible cystoscopy (CSC) and ImmunoCyt followed byPDD in initial diagnosis and CSC followed by WLC infollow-up] had an incremental cost per life-year of over£270,000. The least effective strategy [cytology followedby WLC in initial diagnosis (average cost over 20years £1403, average life expectancy 11.59)] was mostlikely to be considered cost-effective when society’swillingness to pay was less than £20,000 per life-year.No strategy was cost-effective more than 50% of thetime, but four of the eight strategies in the probabilisticsensitivity analysis (three involving a biomarker orPDD) were each associated with a 20% chance of beingconsidered cost-effective. In sensitivity analyses theresults were most sensitive to the pretest probability ofdisease (5% in the base case).iii© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4ContentsList of abbreviations ................................. viiExecutive summary .................................. ix1 Background ............................................... 1Description of health problem ................... 1Current service provision ........................... 6Description of the technologies underassessment .............................................. 92 Definition of the decision problem .......... 15Decision problem ....................................... 15Aim of the review ........................................ 19Structure of the remainder of the report ... 203 Methods for reviewing test performanceand effectiveness ....................................... 21Identification of studies .............................. 21Inclusion and exclusion criteria ................. 22Data extraction strategy ............................. 23Quality assessment strategy ........................ 23Data analysis ............................................... 244 Results – photodynamic diagnosis ........... 27Number of studies identified ..................... 27Number and type of studies included ........ 27Number and type of studies excluded ....... 27Characteristics of the included studies ...... 27Quality of the included studies .................. 30Assessment of diagnostic accuracy ............. 32Recurrence/progression of disease ............. 38Summary – assessment of diagnosticaccuracy and recurrence/progression ofdisease .................................................... 435 Results – biomarkers and cytology .......... 47Number of studies identified ..................... 47Number and type of studies included ........ 47Number and type of studies excluded ....... 47Overview of the biomarkers/cytologychapter ................................................... 47Fluorescence in situ hybridisation .............. 49ImmunoCyt ................................................ 52NMP22 ....................................................... 56Cytology ..................................................... 60Studies directly comparing tests ................. 66Studies reporting combinations of tests ..... 66Summary .................................................... 69© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.6 Assessment of cost-effectiveness ............ 73Economic model for initial diagnosis andfollow-up of bladder cancer ................... 73Results ........................................................ 90Summary of results ..................................... 1087 Assessment of factors relevant to theNHS and other parties ............................. 119Factors relevant to the NHS ....................... 119Factors relevant to other parties ................. 1198 Discussion .................................................. 121Statement of principal findings ................. 121Strengths and limitations of theassessment .............................................. 126Uncertainties .............................................. 126Cost-effectiveness analysis .......................... 1309 Conclusions ............................................... 135Implications for service provision .............. 135Suggested research priorities ..................... 136Acknowledgements .................................. 139References ................................................. 141Appendix 1 Search strategies .................... 153Appendix 2 PDD quality assessmentchecklist (QUADAS tool) ............................ 159Appendix 3 PDD quality assessmentchecklist (RCTs) .......................................... 161Appendix 4 Photodynamic diagnosis(PDD) included studies .............................. 163Appendix 5 Photodynamic diagnosisexcluded studies ......................................... 167Appendix 6 Characteristics of thePDD diagnostic studies .............................. 169Appendix 7 Quality assessment resultsfor the individual PDD studies ................... 177Appendix 8 Studies of PDD versus WLCincluded in pooled estimates for patientandbiopsy-level analysis and also thosereporting stage/grade ................................. 181v

ContentsAppendix 9 PDD and WLC test performancefor detecting bladder cancer, results tablewith 2 × 2 data ............................................. 185Appendix 10 Biomarker/cytologyincluded studies .......................................... 201Appendix 11 Biomarker/cytologyexcluded studies ......................................... 207Appendix 12 Characteristics of thebiomarker and cytology studies .................. 219Appendix 13 Quality assessment resultsfor the biomarker and cytology studies ...... 231Appendix 14 Studies of biomarkersincluded in pooled estimates for patientlevelanalysis and also those reportingspecimen and stage/grade .......................... 235Appendix 15 Biomarker and cytologytest performance for detecting bladdercancer, results table with 2 × 2 data ............. 245Appendix 16 Cut-offs for a positive testused in studies reporting FISH .................. 301Appendix 17 Model structure ................... 303Appendix 18 Summary of studies reportingprognosis and all-cause mortality rates forthe UK ........................................................ 305Appendix 19 Results of cost–consequenceanalysis ....................................................... 309Appendix 20 Cost-effectivenessacceptability curves for the eight strategiesfor changes in the incidence rate(base case = 5%) .......................................... 313Appendix 21 Cost-effectivenessacceptability curves for changes to theperformance of flexible cystoscopy(base-case flexible cystoscopy is the sameas white light rigid cystoscopy) ................... 315Appendix 22 Cost-effectivenessacceptability curves for changes to the relativerisk (RR) of progression of bladder cancer forno treatment of bladder cancer compared withtreatment of bladder cancer(base-case RR = 2.56) ................................. 317Appendix 23 Cost-effectiveness acceptabilitycurves for the eight strategies for changes inthe relative risk (RR) for recurrence comparingPDD with WLC (base-case RR = 1) ............. 319Appendix 24 Cost-effectiveness acceptabilitycurves for the eight strategies for changesin the relative risk (RR) for progressioncomparing PDD with WLC(base-case RR = 1) ...................................... 321Appendix 25 Cost-effectiveness acceptabilitycurves for the eight strategies for changesin the discount rate (base-case discountrate = 3.5%) ................................................ 323Appendix 26 Cost-effectiveness acceptabilitycurves for the eight strategies for changes inproportions in the risk groups for non-invasivedisease (base case: proportion in low-riskgroup is 0.1 and proportion is high-riskgroup is 0.45) ............................................. 325Appendix 27 Cost-effectiveness acceptabilitycurves for the eight strategies for changes inthe starting age and time horizon .............. 327Appendix 28 Cost-effectiveness acceptabilitycurves for the eight strategies when WLCis replaced by PDD in follow-up for eachstrategy ....................................................... 329Appendix 29 Cost-effectiveness acceptabilitycurves for the eight strategies when qualityof life measures are incorporated to producequality-adjusted life-years ........................... 331Health Technology Assessment reportspublished to date ...................................... 333Health Technology Assessmentprogramme ............................................... 353vi

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4List of abbreviations5-ALA5-aminolaevulinic acidMRImagnetic resonance imagingAUABAUSAmerican Urological AssociationBritish Association of UrologicalSurgeonsMVACNATmethotrexate, vinblastine,adriamycin, cisplatinN-acetyltransferaseBCGBMCEACCICISCSCCTCTLDOREAUEORTCbacillus Calmette–Guerinbiomarkercost-effectiveness acceptabilitycurveconfidence intervalcarcinoma in situflexible cystoscopycomputerised tomographycytologydiagnostic odds ratioEuropean Association of UrologyEuropean Organisation forResearch and Treatment ofCancerNCRINICENMP22NPVPDDPPIXPPVQALYQoLRCTReBIPNational Cancer ResearchInstituteNational Institute for Health andClinical Excellencenuclear matrix proteinnegative predictive valuephotodynamic diagnosisprotoporphyrin IXpositive predictive valuequality-adjusted life-yearquality of liferandomised controlled trialReview Body for InterventionalProceduresFDAFood and Drug AdministrationRRrelative riskFISHGCGSTHALHRGHSROCICERIVPMDTfluorescence in situ hybridisationgemcitabine, cisplatinglutathione S-transferasehexaminolaevulinateHealthcare Resource Grouphierarchical summary receiveroperating characteristicincremental cost-effectivenessratiointravenous pyelographymultidisciplinary teamSIGNSROCTCCTURTURBTWHOWLCWMDScottish IntercollegiateGuidelines Networksummary receiver operatingcharacteristictransitional cell carcinomatransurethral resectiontransurethral resection of bladdertumourWorld Health Organizationwhite light cystoscopyweighted mean differenceAll abbreviations that have been used in this report are listed here unless the abbreviation is wellknown (e.g. NHS), or it has been used only once, or it is a non-standard abbreviation used only infigures/tables/appendices, in which case the abbreviation is defined in the figure legend or in thenotes at the end of the table.vii© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Executive summaryBackgroundBladder cancer is the sixth most common cancerin the UK, affecting more than 10,000 people eachyear. Around 75–85% of patients are diagnosed ashaving non-muscle-invasive disease, which, despitetreatment, has a probability of recurrence at 5years of 31% (95% CI 24% to 37%) to 78% (95% CI73% to 84%). Inspection of the bladder [flexiblecystoscopy using white light (CSC)] facilitatedwith local anaesthesia and voided urine cytology(involving the examination of cells in voidedurine to detect the presence of cancerous cells)are currently the routine initial investigations ofthe bladder in patients with haematuria or othersymptoms suggestive of bladder cancer. If CSCor urine cytology are suspicious, a rigid whitelight cystoscopy (WLC) under general or regionalanaesthesia is performed with transurethralresection of bladder tumour (TURBT) whereapplicable. However, WLC may fail to detect sometumours. Photodynamic diagnosis (PDD) is atechnique that could potentially be used to enhancetumour detection. Also, since the mid-1990s manyurine biomarker tests for detecting bladder cancerhave been developed, including fluorescence insitu hybridisation (FISH), ImmunoCyt and nuclearmatrix protein (NMP22).ObjectivesThis review aims to assess the clinical and costeffectivenessof PDD compared with WLC, andurine biomarkers (FISH, ImmunoCyt, NMP22) andcytology for the detection and follow-up of bladdercancer.MethodsElectronic searches were undertaken to identifypublished and unpublished reports. The databasessearched included MEDLINE, MEDLINE In-Process, EMBASE, BIOSIS, Science Citation Index,Health Management Information Consortium(HMIC) and the Cochrane Controlled TrialsRegister as well as current research registers.The date of the last searches was April 2008. The© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.types of studies considered for test performancewere randomised controlled trials (RCTs), nonrandomisedcomparative studies and diagnosticcross-sectional studies that reported the absolutenumbers of true and false positives and negatives.Only RCTs were considered for studies reportingeffectiveness. Participants had symptoms suspiciousfor bladder cancer or were previously diagnosedwith non-muscle-invasive disease. The testsconsidered were (1) PDD compared with WLCor (2) FISH, ImmunoCyt, NMP22 or cytology,with a reference standard of histopathologicalexamination of biopsied tissue.One reviewer screened the titles and abstracts of allreports identified by the search strategy and dataextracted included full-text studies, with checkingby a second reviewer. Two reviewers independentlyassessed the quality of the diagnostic studies usinga modified version of the QUADAS instrumentand the quality of the effectiveness studies using achecklist adapted from Verhagen and colleagues.The results of the individual studies were tabulatedand sensitivity, specificity, positive and negativelikelihood ratios, and diagnostic odds ratios(DORs) calculated. Separate summary receiveroperating characteristic (SROC) curves werederived for different levels of analysis. Metaanalysismodels were fitted using hierarchicalsummary receiver operating characteristic(HSROC) curves. Summary sensitivity, specificity,positive and negative likelihood ratios and DORsfor each model were reported as median and 95%confidence interval (CI). For studies reportingeffectiveness outcomes meta-analysis was employedto estimate a summary measure of effect, withdichotomous outcome data combined using relativerisk (RR). Results were reported using a fixed-effectmodel in the absence of statistical heterogeneity.An economic model was constructed to assessthe cost-effectiveness of alternative diagnosticand follow-up strategies for the diagnosis andmanagement of patients suspected of havingbladder cancer. The model described care pathwaysfrom initial presentation, through diagnosis andtreatment over a 20-year time horizon. A totalof 26 different strategies were considered in theix

Executive summaryxeconomic model, which represented plausible waysin which the tests might be used for the diagnosisand follow-up of patients with bladder cancer. Ofthese 26, eight strategies that appeared to performbest in the deterministic analysis were furtherconsidered in a probabilistic analysis. The clinicaleffectiveness data from the systematic review(summarised below) were incorporated into themodel. In the base-case analysis it was assumedthat the underlying risk of disease within thetarget population was 5%. Costs for treatments andinterventions with strategies were derived from theliterature review in the UK setting, in particularNHS resources. The mean cost per test for PDDwas £1371, WLC £937, CSC £441, cytology £92,NMP22 £39, ImmunoCyt £54 and FISH £55.TURBT cost from £2002 to £2436 dependingupon whether it was assisted by WLC or PDDrespectively. Additional subsequent treatmentswere also included, which were based upon thosetypically adopted within the UK NHS. A cost–utilityanalysis was not possible as part of the base-caseanalysis because of a lack of relevant utility data.Hence, cost-effectiveness (life-years, cases of truepositives) and cost–consequence analyses wereconducted. Sensitivity analyses were conductedto assess the uncertainties in estimates andassumptions.ResultsA total of 27 studies enrolling 2949 participantsreported PDD test performance. In the pooledestimates for patient-level analysis, based on directevidence, PDD had higher sensitivity than WLC(92%, 95% CI 80% to 100% versus 71%, 95% CI49% to 93%) but lower specificity (57%, 95% CI36% to 79% versus 72%, 95% CI 47% to 96%). Inthe pooled estimates for biopsy-level analysis, basedon direct evidence, PDD also had higher sensitivitythan WLC (93%, 95% CI 90% to 96% versus 65%,95% CI 55% to 74%) but lower specificity (60%,95% CI 49% to 71% versus 81%, 95% CI 73% to90%).Across studies, the median sensitivities (range)of PDD and WLC for detecting lower risk, lessaggressive tumours were broadly similar forpatient-level detection [92% (20% to 95%) versus95% (8% to 100%)], but sensitivity was higherfor PDD than for WLC for biopsy-level detection[96% (88% to 100%) versus 88% (74% to 100%)].However, for the detection of more aggressive,higher risk tumours the median sensitivity ofPDD for both patient-level [89% (6% to 100%)]and biopsy-level [99% (54% to 100%)] detectionwas higher than those of WLC [56% (0% to100%) and 67% (0% to 100%) respectively]. Thesuperior sensitivity of PDD was also reflected in thedetection of carcinoma in situ (CIS) alone, bothfor patient-level [83% (41% to 100%) versus 32%(0% to 83%)] and biopsy-level [86% (54% to 100%)versus 50% (0% to 68%)] detection.Four RCTs enrolling 709 participants comparingPDD with WLC reported effectiveness outcomes.The use of PDD at TURBT resulted in fewerresidual tumours at check cystoscopy (pooledestimate RR 0.37, 95% CI 0.20 to 0.69) and longerrecurrence-free survival (pooled estimate RR 1.37,95% CI 1.18 to 1.59) compared with WLC. Theadvantages of PDD at TURBT in reducing tumourrecurrence (pooled estimate RR 0.64, 95% CI 0.39to 1.06) and progression (pooled estimate RR 0.57,95% CI 0.22 to 1.46) in the longer term were lessclear.A total of 71 studies reported the performanceof biomarkers (FISH, ImmunoCyt, NMP22) andcytology in detecting bladder cancer. In total,14 studies enrolling 3321 participants reportedon FISH, 10 studies enrolling 4199 participantsreported on ImmunoCyt, 41 studies enrolling13,885 participants reported on NMP22 and 56studies enrolling 22,260 participants reported oncytology. In the pooled estimates, based on indirectevidence, sensitivity was highest for ImmunoCytand lowest for cytology. FISH (76%, 95% CI 65%to 84%), ImmunoCyt (84%, 95% CI 77% to 91%)and NMP22 (68%, 95% CI 62% to 74%) all hadhigher sensitivity than cytology (44%, 95% CI 38%to 51%). However, cytology had higher specificity(96%, 95% CI 94% to 98%) than FISH (85%, 95%CI 78% to 92%), ImmunoCyt (75%, 95% CI 68% to83%) or NMP22 (79%, 95% CI 74% to 84%).Cost-effectivenessAlthough the differences in outcomes and costsbetween these strategies appear to be small, thedecision about which strategy to adopt dependsupon society’s willingness to pay for additionalgain. The most effective strategy in terms of truepositive cases (44) and life-years (11.66) was astrategy of CSC and ImmunoCyt followed byPDD in initial diagnosis and CSC followed byWLC in follow-up. This strategy had, however, anincremental cost per life-year of over £270,000.The least effective strategy was cytology followedby WLC in initial diagnosis and follow-up (totalaverage cost over 20 years = £1403 and average

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4life expectancy = 11.59). This strategy was mostlikely to be considered cost-effective when society’swillingness to pay was less than £20,000 per lifeyear.Over most of the ranges of willingness to payvalues there appeared to be no strategy that wouldhave a likelihood of being cost-effective more than50% of the time, but four of the eight strategiesincluded in the probabilistic sensitivity analysiswere each associated with an approximately 20%chance of being considered cost-effective. Threeof these four strategies involved the use of abiomarker or PDD.Sensitivity analysesThe sensitivity analyses indicated that the orderof the least to the most costly strategies remainedthe same when discount rates, RR rates andperformance of CSC were changed. The resultswere most sensitive to the pretest probability ofdisease (5% in the base case). At a 1% probabilityit is most likely that the least costly (and leasteffective) strategy of cytology followed by WLCfor both diagnosis and follow-up would be costeffective.At a 20% prevalence the more effectivestrategies (in terms of diagnostic performance) aremore likely to be worth their increased cost.specificity, than cytology in detecting bladdercancer. A urine biomarker test such as ImmunoCytcould potentially replace some cytology testsif higher sensitivity (fewer false negatives) isconsidered more important than higher specificity(fewer false positives). However, if higher specificityis considered more important then cytology wouldremain the test of choice.Linking diagnostic performance to long-termoutcomes required a number of assumptions tobe made about the structure of the economicmodel and its parameters. Some assumptionswere based on non-UK study data; it is unclearwhether such data are applicable to the UK setting.One assumption concerned starting age and thelength of time over which the benefits from adiagnostic strategy may accrue. In the base-caseanalysis a time period of 20 years and startingage of 67 years were used, although the impact ofshorter time horizons and older starting age wereexplored in the sensitivity analyses. When eitherthe time horizon was reduced or the starting agewas increased, the incremental cost per life-yearincreased as the costs of initial diagnosis andtreatments were not offset by survival and life-yeargains.DiscussionPDD has higher sensitivity (fewer false negatives)than WLC and so will detect cases of bladdercancer missed by WLC, but its lower specificitywill result in more false positives. The advantagesof PDD’s higher sensitivity in detecting bladdercancer overall, and also more aggressive, higherrisk tumours, have to be weighed against thedisadvantages of a higher false-positive rate, whichleads to additional, unnecessary biopsies of normaltissue being taken and potentially additionalunnecessary investigations being carried out andthe resulting anxiety caused to patients and theirfamilies.In the four studies reporting effectivenessoutcomes, such as tumour recurrence, theadministration of single-dose adjuvantchemotherapy following TURBT, which canreduce recurrence rates by up to 50% in the first2 years, varied, making it difficult to assess whatthe true added value of PDD might be in reducingrecurrence rates in routine practice.Based on indirect comparisons, all threebiomarkers had higher sensitivity, but lower© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.ConclusionsImplications for service provisionPDD has higher sensitivity than WLC in detectingbladder cancer and is better at detecting moreaggressive, higher risk tumours, including CIS, buthas lower specificity. Based on limited evidence, theuse of PDD at TURBT compared with WLC resultsin fewer residual tumours at check cystoscopyand longer recurrence-free survival, whereas theadvantages of PDD at TURBT in reducing tumourrecurrence and progression in the longer term areless clear. In the pooled estimates ImmunoCyt hadthe highest sensitivity and cytology had the highestspecificity, with all three biomarkers having highersensitivity, but lower specificity, than cytology.Taking into account the assumptions made inthe model, the strategy of CSC and ImmunoCytfollowed by PDD in initial diagnosis and CSCfollowed by WLC in follow-up is likely to be themost costly and the most effective (£2370 perpatient and 11.66 life-years). There appeared to beno strategy that would have a likelihood of beingcost-effective more than 50% of the time overmost of the ranges of willingness to pay values.Nevertheless, strategies involving biomarkers and/xi

Executive summaryor PDD provide additional benefits at a cost thatsociety might be willing to pay. Strategies involvingcytology are unlikely to be considered worthwhile.Strategies that replaced WLC with PDD providedmore life-years but it is less clear whether theywould be worth the extra cost.Recommendations for researchFurther research is required in the following areas:• RCTs including economic evaluationscomparing PDD with rigid WLC at TURBTplus adjuvant immediate single-doseintravesical chemotherapy in patientsdiagnosed with bladder tumours at CSC.• Diagnostic cross-sectional studies comparingFISH with ImmunoCyt, NMP22 BladderChekpoint of care test and voided urine cytologywithin the setting of the British Association ofUrological Surgeons and the Renal Associationdiagnostic algorithm for the diagnosis ofpatients with haematuria. Data producedshould be incorporated into an economicevaluation.• Studies to collect health state utilities areneeded. These may come from furtherprospective studies or as part of future RCTs.• The trade-off between process of care andshort-term (diagnostic outcomes) and longertermoutcomes needs to be explored usingrecognised preference elicitation methodologyin a way that can be incorporated into futureeconomic evaluations.• The impact that an incorrect diagnosis (falsenegativeresult) has on patients either atdiagnosis or at follow-up in terms of futuresurvival, quality of life and costs.xii

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Chapter 1BackgroundDescription of healthproblemIntroductionBladder cancer, or more precisely malignantneoplasm of the bladder, 1 is a disease in whichthe cells lining the urinary bladder lose theability to regulate their growth and start dividinguncontrollably. 2 This abnormal growth resultsin a mass of cells that form a tumour. Peoplewith a suspicion of bladder cancer mainlypresent with urinary symptoms including grosshaematuria, microscopic haematuria and urinarytract symptoms. Bladder cancers can be broadlycategorised into two main groups dependingupon their extent of penetration into the bladderwall: non-muscle invasive and muscle invasive.The majority of diagnosed patients (75–85%)present with non-muscle-invasive disease, which asdescribed in the next subsection is characterisedby a probability of recurrence at 5 years from 31%(95% CI 24% to 37%) to 78% (95% CI 73% to 84%)despite treatment. 3 The remaining cancers aremuscle invasive and/or metastatic.Aetiology, pathology andprognosisAetiologyThe aetiology of bladder cancer appears to bemultifactorial, with environmental and geneticfactors as well as endogenous molecular factorshaving potential roles. The risk of developingbladder cancer before the age of 75 years is2–4% for men and 0.5–1% for women. 4 Cigarettesmoking and specific occupational exposures arethe main known risk factors for bladder cancer. 5In Europe it is estimated that up to half of bladdercancer cases in men and one-third of cases inwomen are caused by cigarette smoking. 6,7Occupational exposure to chemicals in Europeaccounts for up to 10% of male bladder cancers.Most carcinogens have a latent period of 15–20years between exposure and the developmentof tumours. The proportion may be higher incountries with less well-regulated industrialprocesses. Bladder cancer has an importantplace in the history of occupational disease. In© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.1895, Rehn reported cases of bladder cancer in aGerman aniline dye factory. It was then recognisedthat aromatic amines and polycyclic aromatichydrocarbons, by-products of the catabolic process,were the key aetiological factors. Aromatic amineswere widely used in the manufacture of dyesand pigments for textiles, paints, plastics, paperand hair dyes, and in drugs and pesticides andin the rubber industry. In 1953, bladder cancerbecame a prescribed industrial disease in the UK. 8Occupational studies of hairdressers have producedconflicting results. Within the EU, the ScientificCommittee on Cosmetic Products and Non-FoodProducts aims to set up a ‘high-risk’ permanent andsemi-permanent register of hair dye formulations.Several dietary factors have been related to bladdercancer, but the results of different studies havebeen controversial. A meta-analysis 9 of 38 articlessupported the hypothesis that vegetable andfruit intake reduced the risk of bladder cancer.Phenacetin, chlornaphazine and cyclophosphamidealso increase the risk of bladder cancer. 10 Incomparison to other carcinogenic agents, thelatency period is relatively short. Acrolein, ametabolite of cyclophosphamide, is responsiblefor the ninefold increased risk of bladder cancerassociated with cyclophosphamide. In addition,chronic infection by Schistosoma haematobium is acause of squamous cell carcinoma of the bladder.Patients treated with pelvic radiotherapy forcervical and prostate cancers also have an increasedrisk of developing bladder cancer. 11,12Drug- and carcinogen-metabolising enzymes areimportant in the processing of lipophilic chemicalsto products that are more water-soluble and canbe excreted. These enzyme systems are partlycontrolled by genetic polymorphism. In the liver,chemicals are oxidised by the cytochrome P450superfamily and detoxified by N-acetylation,predominantly by N-acetyltransferases (NAT).Aromatic amines are usually detoxified byNAT2. NAT2 slow acetylator genotypes are atincreased risk of bladder cancer [relative risk(RR) 1.4], and this may be especially true insmokers. 13 Approximately 50% of Caucasians and25% of Asians are slow acetylators. GlutathioneS-transferase (GST) is the product of the GSTM11

Backgroundgene and is involved in the detoxification ofpolyaromatic hydrocarbons. Approximately 50% ofCaucasians and Asians have a homozygous deletionof the GSTM1 gene, which is associated with aRR of 1.4. 14 There is no clear evidence that theunderlying pathogenesis of bladder cancer differsby gender. 10PathologyBladder cancer is a disease in which the cells liningthe urinary bladder lose the ability to regulatetheir growth and start dividing uncontrollably.This abnormal growth results in a mass of cells thatform a tumour. The most common type of bladdercancer is transitional cell carcinoma (TCC), whichaccounts for more than 90% of bladder cancersin the UK; other forms of bladder cancer includesquamous carcinoma, adenocarcinoma (urachaland non-urachal), small cell carcinoma, sarcomaand lymphoma. TCC, also known as urothelialcarcinoma, arises from changes in the urothelialcells that line the bladder, ureters, renal pelvis andproximal urethra, although TCC is approximately50 times more common in the bladder thanin other parts of the urinary tract. 15 The 2002TNM staging system of the International Unionagainst Cancer (UICC) 2002 is the most recentpathological staging system (Table 1). 16 About 25%of newly diagnosed TCCs of the bladder are muscleinvasive (T2–T4); the remainder are non-muscleinvasive, either papillary (70%) or a flat lesion ofthe urothelium termed carcinoma in situ (CIS)(5%).For more than three decades, the preferredgrading system in the UK for bladder TCC hasbeen the World Health Organization (WHO) 1973classification, 17 which has been repeatedly validatedTABLE 1 International Union against Cancer (UICC) 2002 TNM staging system2Primary tumour (T) Regional lymph nodes (N) Distant metastasis (M)TX Primary tumour cannot be assessed NX Regional lymph nodes cannot beassessedT0 No evidence of primary tumour N0 No regional lymph nodemetastasisTa Non-invasive papillary carcinoma N1 Metastasis in a single lymphnode, 2 cm or less in greatestdimensionTis Carcinoma in situ: ‘flat tumour’ N2 Metastasis in a single lymphnode, more than 2 cm but notmore than 5 cm in greatestdimension; or multiple lymphnodes, none more than 5 cm ingreatest dimensionT1T2pT2apT2bT3pT3apT3bT4T4aT4bTumour invades subepithelialconnective tissueTumour invades muscleTumour invades superficial muscleTumour invades deep muscleTumour invades perivesical tissueAs for T3 – microscopicallyAs for T3 – macroscopicallyTumour invades any of thefollowing – prostate, uterus, vagina,pelvic wall, abdominal wallTumour invades prostate, uterus,vaginaTumour invades pelvic orabdominal wallN3Metastasis in a lymph node,more than 5 cm in greatestdimensionMXM0M1Distant metastasis cannotbe assessedNo distant metastasisDistant metastasis

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4and shown to be of clinical relevance for treatmentand prognosis. WHO 1973 divides TCC into threegrades on the basis of cytological and architecturaldisorder, grade 1 being well differentiated, grade2 moderately differentiated and grade 3 poorlydifferentiated. WHO 2004 is the latest version ofthe bladder TCC classification. Current reportingguidelines recommend providing the urologistwith both classifications. The main differencesare two grades of carcinoma (high grade and lowgrade) and the introduction of the term papillaryurothelial neoplasm of low malignant potential(PUNLMP) to replace the best differentiated grade1 tumours, avoiding the term carcinoma. However,there has been considerable resistance in the UK toadopting the WHO 2004 classification, which wasnot prospectively validated before its introductionand which has subsequently not demonstratedeither improved reproducibility or clinicalrelevance over WHO 1973. 18 In this report we willtherefore only quote the WHO 1973 classification.PrognosisThe natural history of treated non-muscleinvasivebladder cancer (Ta/T1/CIS), a group ofheterogeneous cancers, can be summarised as anyof the following:• no further recurrence• local recurrence, which can occur on a singleoccasion or on multiple occasions; it caninvolve single or multiple tumour recurrences,but recurrent tumours are usually of the samestage and grade as the primary tumour• local progression – an increase in local stageover time to muscle invasion or the appearanceof distant metastases and subsequent death.On average, non-muscle-invasive bladder cancerhas a probability of recurrence at 5 years from 31%(95% CI 24% to 37%) to 78% (95% CI 73% to 84%)and of progression of between 0.8% (95% CI 0% to1.7%) and 45% (95% CI 35% to 55%) after initialtreatment. 3 The rates of recurrence and progressionvary depending upon the stage, grade and numberof tumours at the time of first presentation. Ofthe newly diagnosed non-muscle-invasive bladdertumours, approximately 30% are multifocal atpresentation. There is little information on thepredictive role of environmental and genetic riskfactors on tumour recurrence, progression andmortality. Tumours are most likely to recur within5 years after transurethral resection of bladdertumour (TURBT), 19 and therefore patients areclosely monitored for recurrence following theirinitial presentation and treatment. According© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.to the European Organisation for Research andTreatment of Cancer (EORTC), the risk factorsrelating to recurrence and progression includethe number of tumours present at diagnosis, therecurrence rate in the previous period, the tumoursize (larger tumours being associated with greaterrisk), stage, grade and the presence of concomitantCIS. 20 The poor prognosis of T1G3 TCC is welldescribed; 50% progression rate if associated withconcomitant CIS. 21 If primary CIS is diffuse, 50% ofthese patients die of metastatic TCC within a yearor two if maintenance intravesical immunotherapywith bacillus Calmette–Guerin (BCG) is notinstituted. Once the tumour has invaded thedetrusor muscle, 50% of patients have occultmetastatic disease at presentation.EpidemiologyBladder cancer is the sixth most common cancerin the UK. 22 Bladder cancer is the most frequentlyoccurring tumour of the urinary system andaccounts for 1 in every 28 new cases of cancerdiagnosed each year in the UK. During the lastthree decades there has been a gradual decreasein the incidence of bladder cancer (Figure 1). 22However, changing trends in the incidence ofbladder cancer over time are difficult to interpretbecause of different and changing classificationsand coding practices of the condition. 5Incidence and prevalenceBladder cancer is the fourth most common cancerin men and the tenth most common in women inthe UK. 22 In 2005, the estimated male and femalecrude incidence rates of bladder cancer were 24.6and 9.3 per 100,000 population with 6091 and2403 new cases, respectively, in England, and 43.0and 17.2 per 100,000 population with 619 and 260new cases, respectively, in Wales (Table 2). 22Although the overall incidence of bladder cancerin the UK has remained much higher in men thanin women in the last five decades, it has shown aslow decrease between 1993 and 2005 (Figure 1)following a rapid rise between 1971 and 1993. 22,23In addition, in England and Wales, the prevalenceof bladder cancer increased by 57% between 1971and 1998, particularly in women. 23Variation in incidence by ageThe mean age at which bladder cancer isdiagnosed in the UK is 71.3 years. The incidenceand mortality rate of bladder cancer rapidlyincrease with increasing age (Figures 2 and 3).3

BackgroundRate per 100,000 population353025201510501993199419951996199719981999200020012002200320042005MalesPersonsFemalesYear of diagnosisFIGURE 1 Age-standardised (European) incidence rates of bladder cancer by sex, UK, 1993–2004.TABLE 2 Number of new cases and rates of bladder cancer in the UK, 2005England Wales Scotland N. Ireland UKCasesMale 6091 619 468 132 7310Female 2403 260 247 58 2968Total 8494 879 715 190 10,278Crude rate per 100,000 populationMale 24.6 43.0 19.1 15.6 24.8Female 9.3 17.2 9.4 6.6 9.7Total 16.8 29.8 14.0 11.0 17.1Age-standardised rate (European) per 100,000 populationMale 19.6 31.6 15.5 15.0 19.8Female 5.7 10.1 5.6 4.4 5.9Total 11.7 19.6 9.8 9.1 11.9Source: Cancer Research UK. 22Number of cases1600120080040000–45–910–1415–1920–2425–2930–3435–3940–4445–4950–5455–5960–6465–6970–7475–7980–8485+3002001000Rate per 100,000 populationMale casesFemale casesMale ratesFemale ratesAge at diagnosis4FIGURE 2 Numbers of new cases and age-specific incidence rates of bladder cancer by sex, England and Wales, 2005.

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Bladder cancer commonly occurs in older peopleand is rare in people under 50 years of age.Variation in incidence by deprivation andgeographyIn the UK the incidence of bladder cancer alsovaries according to socioeconomic status andgeographical area. Data from Cancer ResearchUK 22 show that the incidence is likely to be slightlyincreased in areas of deprivation, with the lowestincidence found in the most affluent groups.Geographical patterns of bladder cancer incidenceare difficult to interpret because of differences inthe way in which bladder tumours are classifiedbetween cancer registries, for example differencesbetween UK and Northern Ireland. Suchdifferences also hinder reliable internationalcomparisons.Impact of the health problemSignificance for patients in terms of illhealthAlthough most non-muscle-invasive bladdercancers are unlikely to be life-threatening theyare associated with high recurrence and variableprogression rates, which result in an impairedquality of life. Untreated bladder cancer isassociated with significant morbidity, such ashaematuria, dysuria, irritative urinary symptoms,urinary retention, incontinence, ureteralobstruction and pelvic pain. In addition to thephysical damage caused, bladder cancer also has asevere effect on work status, sexual life and mentalhealth. A consequence of our population livinglonger will be an increased incidence of bladdercancer with resulting increased morbidity andmortality. At the same time, less smokers in thepopulation may slow the rate of increase.In the UK and also in other countries, unlike othercommon cancers, men with bladder cancer haveconsistently higher survival rates than womenand this also extends to stage-specific survival.Although men seem to be diagnosed at a slightlyearlier stage than women, the reasons for this malesurvival advantage remain unclear.Patients with non-muscle-invasive tumours have5-year survival rates of between 80% and 90%. 5However, patients with muscle-invasive bladdercancer have 5-year survival rates of less than50%, because, although radical treatment dealseffectively with locally invasive disease, manypatients die from metastatic disease, which mayhave been micrometastatic at presentation. 24 Earlydetection while the tumour is still at a non-muscleinvasivestage is therefore very important.Patients with early bladder cancer may fall intoone of three different groups: (1) those with lowriskdisease in whom the main risk is recurrentlow-risk disease with a small chance of everdying of bladder cancer; (2) those with high-risksuperficial disease in whom there is a high chanceof disease progression and subsequent deathfrom bladder cancer; and (3) those with muscleinvasivedisease in whom there is imminent riskof death from bladder cancer. In groups 2 and3, inaccurate diagnosis/follow-up may have lifethreateningconsequences, whereas in group 1 themain impact of follow-up is to prevent morbidityrather than mortality. Therefore the clinical needsNumber of cases80070060050040030020010025020015010050Rate per 100,000 populationMale deathsFemale deathsMale ratesFemale rates000–45–910–1415–1920–2425–2930–3435–3940–4445–4950–5455–5960–6465–6970–7475–7980–8485+Age at diagnosisFIGURE 3 Numbers of deaths from and age-specific mortality rates of bladder cancer by sex, UK, 2006.5© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Background6of these groups differ with respect to diagnosticperformance.Significance for the NHSBladder cancer is considered to be the mostexpensive cancer in terms of lifetime and treatmentcosts because of the high recurrence rates. Ahigher incidence of non-muscle-invasive disease,longer survival requiring lifelong surveillance andtreatment of recurrences are some of the reasonsfor the higher cost of non-muscle-invasive diseasecompared with muscle-invasive bladder cancers.However, annual research fund allocation forbladder cancer from the National Cancer ResearchInstitute (NCRI) UK is less than those for othercancers.Current service provisionDiagnosisHaematuria is presence of blood in the urine andis the most common symptom of bladder cancer.Bladder cancer is detected in approximately 10%of patients with gross haematuria and 3–5% ofthose with microscopic haematuria aged over 40years. 25,26 Less commonly, individuals may notedisturbance in their urinary habits includingcomplaints of dysuria (painful urination), increasedfrequency, urgency of urination, failed attemptsto urinate and urinary tract infection. Thesesymptoms can raise suspicion of diffuse CIS.Other symptoms that may be attributed to a massin the bladder or ureteral obstruction are likelyto indicate that bladder cancer may be muscleinvasivedisease. 5,24,27History, physical examination andradiologyThe clinical workup for potential bladder cancershould start with a history and a complete physicalexamination with careful attention to potential riskfactors, such as the patient’s smoking history andoccupation. Clinicians must look for cancer in allareas of the urinary tract. Most haematuria clinicsin the UK perform an ultrasound of the uppertracts and kidney, ureter and bladder radiography.In some centres, intravenous pyelography (IVP) isalso performed routinely; in others, computerisedtomography (CT) urography has replacedultrasound and IVP in this setting.Cystoscopy and pathologyIn many centres, voided urine for cytologicalanalysis is usually collected before flexiblecystoscopy. Flexible cystoscopy is an invasiveprocedure in which an endoscope is passed withinthe urethra, prelubricated with local anaestheticgel. Its purpose is to evaluate the urethra and tolook for tumours and irregularities in the bladdersuch as red patches (which may prove to be CISon biopsy), diverticula and trabeculations. A urineculture should be performed if dipstick analysissuggests a urinary tract infection.Transurethral resection and/or biopsyIf a bladder tumour is identified on flexiblecystoscopy, arrangements are made for the patientto return as an inpatient for TURBT and/or biopsyunder general anaesthesia. Depending on thelocation of the tumour, resection may be aided onoccasion by muscle paralysis to avoid complicationsarising from an obturator nerve jerk. The exophytictumour is first resected and then a separate deepresection is obtained. Both specimens are sentseparately for histological assessment. Biopsies ofany red areas may also be taken and submitted foranalysis. Haemostasis is then achieved by usinga rollerball electrode followed by insertion of anirrigating catheter. As part of clinical staging, abimanual examination is performed to identify ifthere is a residual mass at the end of the procedure.If a mass is detected, it is noted whether it is mobile(clinical T3) or fixed (clinical T4).Imaging techniquesIf bladder cancer is detected, accurate diseasestaging and grading are critical. There is muchdebate over the role of imaging techniques, suchas magnetic resonance imaging (MRI) and CT, inthe staging of bladder cancer. 27 A staging CT scanof chest, abdomen and pelvis and/or MRI of pelvisare therefore not usually performed in patientswith papillary non-muscle-invasive TCC. The roleof CT in patients with muscle-invasive disease isprimarily to provide extra information on localstaging, lymph node status and visceral metastases.The primary role of MRI in patients with muscleinvasiveTCC is to provide further information onlocal stage.Management of diseaseThe management of non-muscle-invasive bladdercancer is based on: (1) the pathological findings ofthe biopsy specimen, with attention to histologicaltype, grade and depth of invasion; (2) the presenceof associated CIS; (3) the number of tumours; (4)previous recurrence rate if applicable; and (5) sizeof tumour. Depending on these findings, treatmentoptions include cystoscopic follow-up only(either flexible or rigid cystoscopy under general

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4anaesthesia), cystoscopic follow-up and intravesicalchemotherapy and immunotherapy courses orradical cystectomy.The goals of current treatment for patients withnon-muscle-invasive bladder cancer are to preventdisease recurrence or progression to muscleinvasivedisease to avoid loss of the bladder and,ultimately, to enhance survival. The currenttreatment strategies for patients with bladdercancer depend on three main types of bladdercancer, non-muscle-invasive disease, muscleinvasivedisease and metastases, as recommendedin the multidisciplinary team (MDT) guideline. 28Non-muscle-invasive diseaseInitial treatment• TURBT of all malignant tissue is therecommended primary treatment for nonmuscle-invasivedisease and should befollowed as soon as possible (ideally within 6hours, otherwise within 24 hours) by a singleinstillation of intravesical chemotherapy.• Tumours should then be assessed dependingon stage, grade, size, multiplicity and thepresence of recurrence at cystoscopy after 3months:––low risk – patients at low risk of recurrenceand progression have TaG1 TCC or solitaryT1G1 TCC––intermediate risk – those at intermediaterisk have TaG2 TCC or multifocal T1G1TCC––high risk – broadly speaking, patients withTa/T1G3 TCC, CIS or multifocal T1G2TCC are classified as being at high risk ofnot only recurrence but also progression.Follow-up of low- and intermediate-risknon-muscle-invasive bladder cancerFollow-up of non-muscle-invasive disease isby cystoscopy, the frequency and duration offollow-up depending on the risk at presentationand the presence of recurrences. Multiplicity atpresentation and a tumour recurrence at 3 monthshave consistently been shown to be key practicalpredictors of future recurrence, and so manyurologists in the UK tailor their cystoscopic followupof low- and intermediate-risk patients based onthese two factors:(a) If patients have a solitary tumour at diagnosisand no tumour recurrence at 3 months theyare then followed up at 9 months and thenannually for 4 further years. If at the endof this 5-year follow-up period they have© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.remained tumour free they are discharged.During the follow-up visits patients undergoflexible cystoscopy and in some centrescytology and/or biomarker tests. Not allpatients with a tumour recurrence will receiveTURBT; some may have a cystodiathermy andbiopsy.(b) Patients with multiple tumours at presentationand no recurrence at 3 months or a solitarytumour at presentation with recurrence at3 months need more intense follow-up andare followed up every 3 months for the firstyear and annually if they remain tumourfree until 10 years and are then discharged.During the follow-up visits patients undergocystoscopy and in some centres cytology and/or biomarker tests. Those who present with atumour at the follow-up visit undergo eitherTURBT or cystodiathermy and biopsy. Thesepatients may be considered for a course of sixintravesical instillations of mitomycin C orepirubicin.(c) Patients with multiple tumours at presentationand recurrence at 3 months have the highestrisk of recurrence and are followed up every 3months for the first 2 years and then annuallythereafter. They are usually offered a course ofsix intravesical instillations of mitomycin C orepirubicin. Those who present with a tumourat follow-up visits undergo either TURBT orcystodiathermy and biopsy. During the followupvisits patients undergo cystoscopy and insome centres cytology and/or biomarker tests.Cystoscopies in the first 2 years are usuallyunder general anaesthesia using a rigidcystoscope. 29Follow-up of high-risk non-muscleinvasivebladder cancerIf diagnosed with T1G3 TCC, patients areoffered an early re-resection to ensure that thetumour is not muscle invasive. All patients in thisgroup are usually offered an induction course ofsix intravesical BCG instillations followed by amaintenance regimen of a further 21 instillationsover a 3-year period. Some may opt for primaryradical cystectomy. Patients who opt for bladdersparing undergo their first bladder check at 3months. If they remain tumour free they arefollowed up every 3 months for the first 2 years andthen every 6 months thereafter. During the followupvisits patients undergo cystoscopy and in somecentres cytology and/or biomarker tests. Patientsfound to have a non-muscle-invasive recurrenceat 3 months have four options: they can undergocystectomy, have a second induction course of BCG7

Background8and then reassess, have three further instillations ofBCG and then reassess, or have endoscopic control.Muscle-invasive diseaseInitial treatmentOnce again, initial treatment comprises TURBT.If muscle invasion is confirmed on histologicalanalysis, patients undergo CT of the chest,abdomen and pelvis and in some centres MRIscanning of the pelvis. In the absence of metastaticdisease and other significant comorbidity,treatment options for patients with muscleinvasivedisease include radical cystectomy withileal conduit formation, radical cystectomy withformation of a neobladder, or radical radiotherapy.Neoadjuvant systemic chemotherapy is usuallyrecommended before radical cystectomy orradiotherapy.Follow-up• Follow-up after radiotherapy is by regular(usually 6-monthly) cystoscopy. The first checkcystoscopy is usually performed at about 4months post completion of radiotherapy.• Follow-up after cystectomy is by clinicalassessment and CT scanning.• A CT scan should be performed (at around 6months following surgery for most patients)to assess for lymph or local recurrence.Subsequent CT scanning may be requiredin some cases but need not be carried outroutinely.• Non-muscle-invasive recurrences are dealt withendoscopically. Intravesical chemotherapy orBCG should be considered if recurrences aremultiple or frequent.• Non-muscle-invasive recurrences afterradiotherapy are dealt with endoscopically.Intravesical chemotherapy, or in advancedcases salvage cystectomy, should be considered.• Muscle-invasive recurrences after radiotherapyare best dealt with by salvage cystectomy ifthe patient’s condition allows (in other caseschemotherapy may be appropriate).• Recurrence after cystectomy may be treatedwith radiotherapy or chemotherapy.Metastatic diseaseRadiotherapy can provide effective palliation forsymptoms of locally advanced disease such ashaematuria. Chemotherapy may be appropriatein cases of metastatic disease in which thepatient has a good performance status and renalfunction. Treatment is purely palliative andshould be selected according to the patient’sneeds but may include systemic chemotherapywith GC (gemcitabine and cisplatin) or MVAC(methotrexate, vinblastine, adriamycin, cisplatin).Combinations with cisplatin are more effective thanthose without. 30,31 Gemcitabine plus cisplatin hasequivalent survival to MVAC but is much less toxic.Non-transitional cell carcinoma bladdercancerCareful case-by-case management of non-TCCbladder cancer patients is required includingdiscussion by the specialist MDT. Specialisthistopathological review may be required, withconsideration to the fact that the primary tumourmay not be arising from the bladder.Current service costIt is difficult to estimate the current bladder cancerservice cost in the UK because of the variation inpractice in the diagnosis and follow-up of patientsbased on their risk categorisation. It is anticipatedthat the costs of the higher risk patients will begreater than those of the low-risk patients becauseof more follow-up interventions. The total cost oftreatment and 5-year follow-up of patients withbladder cancer diagnosed during 2001–2 was£55.39 million; the total cost of superficial diseasewas £35.25 million and that of invasive diseasewas £20.2 million. The total cost for patientsundergoing radical radiotherapy was over twicethat for those undergoing cystectomy (£8.1 versus£3.6 million) 32 In the USA it is estimated that $1.7billion is spent on bladder cancer. 33An estimate of the current cost to the UK NHScan be generated by using the total cost of eachstrategy (see Tables 39 and 42) and combiningit with the values in Table 2. If it assumed thatthe current practice for diagnosis in the UKis flexible cystoscopy and cytology for initialdiagnosis followed by white light rigid cystoscopy[CSC_CTL_WLC (CSC_WLC)] the cost per lowriskpatient will be £6302.25. Therefore the totalannual cost to the NHS will be £64,765,481. Thereis also evidence that costs are likely to increase withimproved survival because patients need severalcourses of treatment.Variation in services and/oruncertainty about best practiceAll urology departments offer haematuriaclinics and subsequent TURBT if appropriateeither in the same hospital or in a hub hospital.Radiotherapy and systemic chemotherapy areavailable in cancer centres. Radical surgery for

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4prostate and bladder cancer should be providedby teams carrying out a cumulative total of at least50 such operations per annum. These proceduresshould be performed by surgeons performingat least five of either radical cystectomy orprostatectomy each year. 34Relevant national guidelines,including National ServiceFrameworksThe relevant national guidelines are:• National Institute for Clinical Excellence(2002). Improving outcomes in urological cancers.NHS guidance on cancer services 34• National Institute for Clinical Excellence(2003). Laparoscopic cystectomy of the urinarybladder. IPG026 35• Scottish Intercollegiate Guidelines Network(SIGN) (2005). Management of transitional cellcarcinoma of the bladder 36• National Institute for Health and ClinicalExcellence (2007). Intravesical microwavehyperthermia with intravesical chemotherapy forsuperficial bladder cancer. IPG235 37• NHS Pan-Birmingham Cancer Network(2006/2007). Guidelines for the management ofbladder cancer 38• UK National Screening Committee (NSC)(2002). Evaluation of urinary tract malignancy(bladder cancer) screening against NSC criteria 39• British Association of Urological Surgeons(BAUS) Section of Oncology and Uro-oncologyGroup (2007). MDT (multi-disciplinary team)guidance for managing bladder cancer 28• European Association of Urology (EAU) (2009).Guidelines on TaT1 (non-muscle-invasive) bladdercancer 3• European Association of Urology (EAU) (2009).Guidelines on bladder cancer: muscle invasive andmetastatic 40• American National Comprehensive CancerNetwork (NCCN) (2009). NCCN clinical practiceguidelines in oncology. Bladder cancer includingupper tract tumours and urothelial carcinoma of theprostate 24• American Urological Association (AUA) (2007).Guideline for the management of nonmuscle invasivebladder cancer (stages Ta,T1 and Tis). 5Only two of the above guidelines specificallymention photodynamic diagnosis (PDD):The evidence suggests potential benefits fromphotodynamic techniques for patients withsuperficial bladder cancer undergoing initial© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.resection of their tumour. Its role in patientsdeveloping recurrence during followup is lessclear.SIGN (2005) 36The benefit of fluorescence-guided TURBTfor recurrence-free survival was shown inseveral small randomised clinical trials, but itsvalue remains to be proven in improving theoutcome of patients for progression rates orsurvival. The additional costs of the equipmentshould be considered.EAU (2009) 3Various guidelines, including those of the EAUand AUA, recommend the use of voided urinarycytology, both in the diagnosis and surveillance ofnon-muscle-invasive bladder carcinoma. However,there are no equivalent recommendations for theuse of biomarkers. Although the internationalconsensus panel on the use of biomarkers inbladder cancer realised the importance of noninvasivediagnosis and surveillance of non-muscleinvasivedisease, it concluded that, although noneof the non-invasive tests could replace cystoscopy,many markers together with cystoscopy couldimprove the current practice of managing patientswith bladder cancer. 41Description of thetechnologies underassessmentSummary of interventionsPhotodynamic diagnosisPrinciplesFluorescenceFluorescence occurs when a molecule absorbsone colour of light and emits another colour.Essentially, photons of light are absorbed by tissueand excite electrons in the tissue. The electronthen returns to its resting state and the photon isemitted with less energy, i.e. longer wavelength,resulting in a different colour emission.Fluorescence cystoscopy is based on the principlethat specific fluorochromes have increasedaffinity for neoplastic tissue compared withnormal urothelium. When light of an appropriatewavelength is used to look at the surface ofbladder to which the fluorochrome has beenapplied, different signal intensities are given off byneoplastic and non-neoplastic tissue. To minimiseautofluorescence from cellular components suchas collagen, a longpass eye filter is needed. A9

Background10filter allowing only wavelengths > 600 nm wouldbe ideal, but this would result in the image beingvery dark. A compromise is therefore to use a450-nm yellow filter and therefore accept someautofluorescence. This does not affect colourreproduction in the white light mode.Over the last 40 years, several agents have beenevaluated for their ability to improve visualisationof urothelial cancer. These include tetracyclines,fluorescein, methylene blue and syntheticporphyrin compounds. However, these havebeen abandoned because of several side effects,including cutaneous toxicity lasting several weekswith synthetic porphyrins.Photosensitisers5-Aminolaevulinic acid-mediated fluorescencecystoscopy A major breakthrough was thediscovery that 5-aminolaevulinic acid (5-ALA),in a suitable dose, could be safely applied tothe bladder surface and permit detection oftumours by fluorescence without serious adverseeffects. 5-ALA is an initial substrate of hemebiosynthesis. Exogenous application of 5-ALAinduces an accumulation of fluorescent porphyrins,predominantly protoporphyrin IX (PPIX), inepithelial tissue. Using a blue–violet light with awavelength of 450 nm, PPIX appears as fluorescentred whereas normal urothelium appears blue.This is because PPIX accumulates up to 10 timesmore in neoplastic cells than in normal tissue.The mechanism of accumulation of fluorescentPPIX in urothelial cancer is unclear. Severaltheories, including a difference in the metabolicrate of neoplastic tissue, hyperproliferation andinflammation-induced increased permeability toALA, have been proposed. These are supportedby the observations that increased PPIX can bedetected in urothelial hyperplasia, inflammationand granulation tissue. 5-ALA is usuallyadministered intravesically 2–3 hours beforecystoscopy at a dose of 1.5 g. The procedurerequires special endoscopes and a specific lightsource (D-light TM , Karl Storz).Hexaminolaevulinate-mediated fluorescencecystoscopy 5-ALA absorption is limited becauseof its positive electric charge. The esterificationof 5-ALA as hexylester aminolaevulinate makesALA more lipophilic, which enables it to crossthe cell membrane more easily. A consequenceof this is more rapid cellular uptake and higherfluorescence than with ALA. 42 Hexaminolaevulinate(HAL) needs therefore only be administered 1hour before cystoscopy and the dose is typically a85-mg solution of HAL hydrochloride in 50 ml ofphosphate buffered saline (Hexvix ® ).Hypericin-mediated fluorescence cystoscopy Recently,hypericin has been proposed as an additionalphotosensitiser. Hypericin consists of ahydroxylated phenanthroperylenequinone thatis extracted from the Hypericum perforatum plant,which is present in St John’s wort. Within anorganic solution, hypericin produces an intense,prolonged, red fluorescence signal. This is becauseits pigment produces single oxygen species uponexposure to light of an appropriate wavelength.Most studies have used hypericin at a concentrationof 8 μmol/l and instilled it 1–2 hours beforecystoscopy.ProcedureBefore TURBT, a 12F LoFric or two-way urethralcatheter is inserted by a nurse on the ward andintravesical photosensitiser instilled. The catheteris removed immediately. In theatre, under generalor spinal anesthesia, the bladder is first inspectedusing white light rigid cystoscopy. The bladder isthen reinspected using blue–violet light. Normalappearingbladder should appear blue. Normalappearingbladder neck and/or prostate appearred because of tangential views that cause them tobe artefactually red. This, however, acts as a usefulpositive control. Within the bladder, any red areasare considered to be suspicious and require biopsy.The bladder tumour is then resected in white light.A further inspection of the bladder with blue–violetlight will then identify any residual tumour thatmay have been missed on WLC.Equipment• Photosensitiser, e.g. 5-ALA, HAL, hypericin.• Rigid cystoscope with longpass yellow filter forwavelengths > 450 nm.• Fluid light cable – this blocks residual infraredlight and lowers intrinsic autofluorescence;however, a disadvantage is that it cannot beautoclaved.• Switchable bandpass filter – this enables thesurgeon to interchange between white light andblue–violet light without changing cystoscopes.• Xenon lamp – powerful, especially in the bluelight spectra.• Camera controller.• Video monitor.• Colour charge-coupled device (CCD) camera(on chip integration) – this is suitablefor working in low light conditions. The

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4fluorescent image is 10 times less intense thanwhite light; allows increased red light intensity.• Beam splitter cube.Extra personnel involvedUnlike white light cystoscopy, PDD requires theinstillation of a photosensitiser via a urethralcatheter before TURBT. This is usually performedby a nurse on the ward.Procedure time compared withconventional cystoscopyOn the ward, catheterisation and instillation of thephotosensitiser and then removal of the cathetertakes about 15 minutes. In theatre, fluorescenceguidedTURBT takes an extra 10 minutescompared with conventional white light TURBTalone.Urinary biomarkersUrinary biomarkers are molecular substances thatcan be objectively measured in urine and evaluatedas indicators of physiological or disease processesin the urinary tract or in various systems of thebody. In principle, this could act as a source ofvital information for diagnosis, prognosis andpredicting response to therapies. The explosion ofinterest in urinary biomarker research, in particularrelated to bladder cancer, is driven by the fact thatthere is a lack of non-invasive methods of diagnosisand disease surveillance. The current standardof care – endoscopic inspection of the inside ofthe urinary bladder – is not only invasive but canalso miss up to 10% of bladder tumours. 43 Theurinary measurement of biomarkers could providea diagnostic means that could either complementcystoscopy to enhance its performance or replace itas a mode of diagnosis and surveillance.From a methodological perspective, urinarymarkers fall into a few broad groups, in particularsoluble urinary proteins, cell-based biomarkers andnucleic acid biomarkers. As a complete review ofeach specific biomarker is beyond the scope of thischapter, the present study focused on four urinarybiomarkers approved by the US Food and DrugAdministration (FDA) for clinical use in urologicalpractice. These are urinary cytology, nuclear matrixprotein (NMP22), fluorescence in situ hybridisation(FISH) and ImmunoCyt.Place of biomarkers in the treatmentpathwayThere are several potential strategies worthconsidering aimed at making use of urinarybiomarkers in the care pathways of bladder cancer.They could be used:• Alone or as an adjunct to urinary cytology toimprove the detection rate of cancer in highriskpopulations.• To provide a less expensive and more objectivealternative to the urinary cytology test.• To replace or supplement direct cystoscopicsurveillance of non-muscle-invasive bladdercancer. They may also serve to decrease thenumber of invasive procedures, provided thatadequate cancer control is maintained onfollow-up, and thereby reduce the health-carecost and improve the comfort of patients.The critical issue remains the operatingcharacteristics of these markers compared withcystoscopy, the current standard of care. Falsepositiveresults are likely to generate furtherunnecessary investigations in addition to fear andanxiety in patients’ minds; alternatively, falsenegativeresults may prove to be detrimental, suchas progression to muscle invasion.SettingUrinary cytologyUrinary cytology involves examination of cellsfrom the urinary tract under microscopy. Aurinary sample is transported to the laboratoryand cells are retrieved by a conventional cytospinmethod. Cells are examined under a microscopeby a cytopathologist for the presence or absenceof malignant changes using the standardPapanicolaou method. The test is laboratorybased and results are observer dependent withthe potential for inter- and intraobservationalvariation.Nuclear matrix proteinNMP22 is a patented proteomic technologythat has been commercialised by Matritech. Twoproducts are marketed for the diagnosis of bladdercancer, the NMP22 ® Test Kit and the NMP22 ®BladderChek ® Test. The NMP22 BladderChek Testis the only in-office test approved by the FDA forthe diagnosis of bladder cancer. It is a non-invasivetest performed on a single urine sample. Bladdercancer cells release NMP22 protein into urine,which is detected by putting 4–5 drops of urine ona prepared card. A change in colour is consideredas a ‘positive test’ result. The levels of NMP22in urine from healthy individuals are very smallbut can be significantly elevated in patients withurothelial cancers. The test has also been approved11© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Background12by the FDA for point of care use in the diagnosis ofbladder cancer.Fluorescence in situ hybridisationThe basis of this test is the detection of abnormalDNA sequences on chromosomes 3, 7, 17, andthe loss of the 9p21 locus in cancer cells shed intothe urine of patients with bladder cancer. Theretrieved cells from voided urine specimens arefixed on microscopy slides and visualised usinga four-colour, four-probe mixture of DNA probesequences homologous to specific regions on theaforementioned chromosomes. This is a laboratorytest and has been commercialised by Abbott underthe market name of UroVysion Bladder CancerKit (UroVysion Kit).ImmunoCytThe ImmunoCyt test uses a cocktail of threemonoclonal antibodies labelled with fluorescentdyes that bind to two antigens, a mucinglycoprotein (green) and a carcinoembryonicantigen (red), expressed by bladder tumour cellsin urine specimens. A voided urine specimen istransported to the laboratory and cells retrievedfrom it are fixed to a microscope slide. Theantibodies are added to the slide and the stainedslide examined under fluorescent microscopy by acytopathologist.Equipment required and personnelinvolvedUrine cytology requires the support of skilledlaboratory cytotechnicians and cytopathologistswithin pathology laboratories. This means thatresults take longer to obtain and are not availableon the same day. In addition to these requirements,the FISH and ImmunoCyt tests require specifickits and specialised fluorescence microscopes forvisualisation of labelled cancer cells. Also, the FISHtechnique requires a special filter for cell retrieval.The only biomarker test approved for point of carediagnosis of bladder cancer is NMP22 detectionusing the commercially available NMP22 Test Kit.The test provides instantaneous results and canbe performed by medical personnel with minimaltraining.Identification of importantsubgroupsPhotodynamic diagnosis• It is important to distinguish the role offluorescence-guided TURBT for primarytumours from its role in bladder tumourrecurrence. Its role in patients developingrecurrence during follow-up is less clear.• It is important to realise that the use ofdifferent photosensitisers may lead to differentresults in terms of sensitivity and specificity.Biomarkers/cytologyThe diagnostic performance of urinary biomarkerscan be scrutinised in the background of two clinicalsettings: the ability to accurately diagnose bladdercancer in high-risk populations and their potentialto accurately predict recurrences in patients knownto have non-muscle-invasive disease. Urinarybiomarkers can either complement or replacecurrent invasive tests such as cystoscopy. Thesecond clinical scenario in which the diagnosticutility of urinary biomarkers comes under sharpfocus is their ability to perform across all gradesand stages of non-invasive bladder cancer disease.For example, urinary cytology performs well(high sensitivity) in high-grade disease, whereasits performance decreases (low sensitivity) in lowgradedisease – this is why it is not a plausiblereplacement for cystoscopy, both at the point ofdiagnosis and at follow-up in the care pathways ofnon-muscle-invasive bladder cancer disease.Current usage in the NHSPhotodynamic diagnosisIn most UK centres PDD is not available. Moreover,in centres in which the service is available, it isused to a varying extent. In a few centres (less thanfive) it is used routinely for all first-time TURBTs.In others it may be used only during follow-upwhen CIS is suspected, such as a normal-appearingbladder on WLC but positive urine cytology.Two further factors are likely to influence theuptake of PDD within the wider NHS:• Fluorescence cystoscopy has been identified asa new technology that has been signalled bythe NCRI to the National Horizon ScanningCentre for early review.• In 2008 the NHS Technology Adoption Centretook forward a PDD implementation projectinvolving three NHS trusts. The experiencegained from the project will support the widerNHS in overseeing issues associated with theadoption of new technologies.Biomarkers/cytologyAlthough urinary cytology is the most commonurinary biomarker used for the diagnosis andfollow-up of non-muscle-invasive bladder cancerin the NHS, the practice varies across the UK. 44There are few reports of NMP22 being used as adiagnostic biomarker in patients with haematuria

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4from UK centres. 45 The clinical use of FISH andImmunoCyt as urinary markers in patients withbladder cancer has not been reported in the UK.Anticipated costs associatedwith the technologiesThe anticipated costs associated with thetechnologies will depend on the strategies usedin the diagnosis and follow-up of patients. Theaverage unit cost of diagnosing bladder cancerusing PDD is £1371, rigid white light cystoscopy£937, flexible cystoscopy £441, cytology £92.37,NMP22 £39.3, FISH £54.8 and ImmunoCyt £54.8;and the cost of treatment using PDD-assistedTURBT is £2436, WLC-assisted TURBT £2002,mitomycin £73, BCG £89, cystectomy £6856,chemotherapy £50.22, radical radiotherapy £1050and palliative treatment £12,825 (see Chapter 6 fordetails). The modelling results indicate that usingthe most effective strategy (the one with the highestnumber of true positives and the lowest number offalse negatives), which includes either of the twobiomarkers FISH or ImmunoCyt and PDD as theinitial strategy and either FISH or ImmunoCyt withWLC as the follow-up strategy, will cost £5919.28per low-risk patient per year.13© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Chapter 2Definition of the decision problemDecision problemAccurate diagnosis of bladder cancer is crucialfor people who may potentially have the diseaseto allow for early detection and to reduce therisk of tumour recurrence and progression. Theideal test for diagnosis and follow-up of bladdercancer would be non-invasive, highly sensitiveand specific, inexpensive and easy to performand would provide reproducible results. Many ofthe tests meet some, but not all, of these criteria.Currently, a common diagnostic scenario in the UKis that people suspected of having bladder cancerare first examined with flexible cystoscopy andvoided urine cytology, followed by white light rigidcystoscopy-assisted TURBT or biopsies for thoseconsidered positive or suspicious for the disease.However, insufficient sensitivity or specificity of thethree tests can result in the incomplete detection orovertreatment of primary and recurrent disease.As patients are living longer and recurrenceof disease is becoming a major issue there is aneed to identify the most appropriate methodsfor diagnosing patients with bladder cancer andsubsequently following them up. A variety oftests have been developed that have been used asalternatives to, or alongside, existing investigations.As described in Chapter 1, urinary biomarkersfor bladder cancer are non-invasive assay teststhat can detect protein, genetic or chromosomalaberrations, even at early stages of disease. Someare point of care tests whereas others requirelaboratory analysis. These tests are consideredto be attractive and potentially cost-effective asthey may offer the potential to avoid unnecessarycystoscopies and labour-intensive cytology.Biomarkers have the potential to play a role inthe initial diagnosis of patients either in additionto or as a replacement for urine cytology, and inmonitoring during follow-up.PDD has been used alongside rigid cystoscopywith the aim of improving detection of CISand papillary tumours during TURBT, therebypotentially reducing the residual tumour rate at the6-week check following TURBT and consequentlyalso reducing recurrence and progression of© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.disease. PDD has also been described as a safe andstraightforward technique to learn.The following sections provide a description of thecare pathways that show the plausible strategies forthe primary diagnosis and follow-up of people withbladder cancer.Inclusion criteria (see Chapter 3)Key issuesThe key issues to be addressed are:• Can PDD improve detection of bladder cancer(1) at the time of TURBT for newly diagnoseddisease and (2) during follow-up of patientswith non-muscle-invasive disease?• Can PDD reduce recurrence and/or progressionof non-muscle-invasive bladder cancercompared with WLC?• Can urine biomarkers (FISH, ImmunoCyt,NMP22) improve detection of bladdercancer during (1) initial diagnosis of patientssuspected of having bladder cancer and (2)follow-up of patients diagnosed with nonmuscle-invasivedisease?• What is the incremental cost-effectiveness ofPDD during TURBT for newly diagnosed nonmuscle-invasivebladder cancer and duringfollow-up?• What is the incremental cost-effectivenessof biomarkers during the initial diagnosis ofpatients suspected of having bladder cancerand during follow-up of those diagnosed withnon-muscle-invasive disease?Care pathwaysCare pathways describing plausible strategiesfor the initial diagnosis and follow-up of peoplewith bladder cancer were developed. The basiccare pathway was based on discussions with theclinical experts involved in this study and a briefdescription of this is provided within Chapter 1.Initial diagnosis and treatment (Figure 4)The pathway begins with an initial presentationof symptoms or asymptomatic microscopichaematuria and varies in terms of where and when15

Definition of the decision problem16Diagnosis/TURBT TreatmentFollow-upFigure 5Non-muscleinvasive (80%)DischargeCSC negativeCTL negativeBM negativeNo bladder cancerCSCCSC + CTLCSC + BMCSC + CTL + BMRadical cystectomy ±three cycles of chemotherapyMicroscopichaematuriaGrosshaematuriaLowerurinary tractsymptomsMuscleinvasive(T2, T3, T4)(15%)Negative(no bladdercancer)Radiotherapy ±three cycles of chemotherapyWLCPDD(TURBT+mitomycin)CSC and/orCTL and/orBM positiveRecurrenceProgressionCancer: deadPositive(bladder cancer)OtherLow risk(TaG1, solitary TIG1)Intermediate risk(TaG2, multi TIG1)High risk(Ta/TG3, multi TIG2)Muscleinvasive (20%)MetastasesCancer: deadPalliativetreatmentMetastases(5%)Other: deadDeadFIGURE 4 Developed care pathway – initial diagnosis/treatment. BM, biomarkers; CSC, flexible cystoscopy; CTL, cytology.

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4biomarkers and PDD might be used. Patientswho present with either microscopic or grosshaematuria or lower urinary tract symptoms aretested using flexible cystoscopy and cytology.Biomarkers could be used at this point either inaddition to these two tests or instead of cytology.The results of these tests can be either negativeor positive. Patients who have two/three negativeresults are discharged. Discharged patients wholater re-present with similar symptoms go backto the beginning of the care pathway. Patientswith one or more positive results for these tests asoutlined in Table 3 undergo TURBT during whichPDD may be used with the aim of improving thedetection of tumours, thereby potentially reducingthe rate of residual tumours and increasing thedetection of CIS and small papillary tumours.After TURBT is performed for newly diagnosedbladder cancer, the standard UK management isthat the patient also receives a single instillation ofadjuvant intravesical mitomycin C, ideally within6 hours of resection but not later than 24 hoursif possible. Biopsies are taken and the results ofthe histological analysis may be either negativeor positive for bladder cancer. Those who havea negative histology result are then discharged.Discharged patients whose symptoms are notresolved may subsequently re-present at thebeginning of the care pathway. For the purposesof this review, although patients who have anegative bladder cancer test result are consideredas discharged, it is noted that some who initiallyhad a positive result may be at risk of upper tracturothelial cancer or renal cancer and consequentlywill require further tests, and, if positive, treatment.Those patients whose histological results confirmthe presence of bladder cancer are classified intoTABLE 3 Different test resultsCystoscopy Cytology Biomarkers– – –– – +– + –+ – –– + ++ – ++ + –+ + +–, negative; +, positive.© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.muscle-invasive or non-muscle-invasive disease.For those with muscle-invasive disease, treatmentoptions are outlined in Figure 4. Essentially,those amenable to potential cure are offeredeither radical cystectomy with bilateral pelviclymphadenectomy or radiotherapy. Treatmentwith surgery or radiotherapy is usually preceded bythree cycles of systemic neoadjuvant cisplatin-basedchemotherapy. The rationale for chemotherapyis that over 50% of patients with muscleinvasivedisease have occult metastatic disease atpresentation. It is noted that practice at individualcentres may vary. The decision for cystectomy orradiotherapy is primarily based on patient choiceand medical fitness. The presence of concomitantCIS and upper tract dilatation are also factors thatfavour cystectomy. For patients with more advancedmetastatic disease, the treatment is palliative.Follow-up of patients with non-muscleinvasivebladder cancer (Figure 5)The key factors increasing the risk of recurrenceand progression in patients with non-muscleinvasivebladder cancer are: (1) tumour multiplicity,(2) greater tumour diameter, (3) previousrecurrence rate, (4) higher T-stage, (5) concomitantCIS and (6) higher histological grade. A briefsummary is provided in the following sectionsand a further short review on the managementof bladder cancer, required for the description ofthe model structure, is provided in Chapter 6 (seeModel structure, Markov model).High riskBroadly speaking, patients with Ta/T1G3 TCC,CIS or multifocal T1G2 TCC are classified asbeing at high risk of not only recurrence but alsoprogression. If diagnosed with T1G3 TCC theyare offered an early re-resection to ensure thatthey are not muscle invasive. All patients in thisgroup are usually offered an induction course ofsix intravesical BCG instillations followed by amaintenance regimen of a further 21 instillationsover a 3-year period. Some may opt for primaryradical cystectomy. Patients who opt for bladdersparing undergo their first bladder check at 3months. If they remain tumour free they arefollowed up every 3 months for the first 2 years andthen every 6 months thereafter. During the followupvisits, patients undergo cystoscopy and in somecentres cytology and/or a biomarker test. Patientsfound to have a non-muscle-invasive recurrencehave four options: they can undergo cystectomy,have a second induction course of BCG and thenreassess, have three further instillations of BCGand then reassess, or receive endoscopic control.17

Definition of the decision problem18First follow-up visit Second follow-up visitAnnually for 4 yearsDischarge if no tumourNo recurrence9 monthsTURBTCSCCTLBMSolitary at presentation and no recurrence at 3 monthsRecurrenceCystodiathermy and biopsy3 monthly for first year, andthen annually for 10 yearsLow riskDischarge if no tumourNo recurrenceCSCCTLBM3 monthsMultiple at presentation and no recurrence at 3 monthsOr solitary at presentation and recurrence at 3 monthsCSCCTLBM3 monthsTURBTRecurrenceCystodiathermy and biopsyIntermediate risk3 monthly for first 2 years and then annuallyNo recurrenceCSCCTLBM3 monthsMultiple at presentation and recurrence at 3 monthsTURBTNon-muscleinvasive(80%)RecurrenceCystodiathermy and biopsy3 monthly for first 2 years and then 6 monthlyNo recurrenceCystectomySecond induction of BCG then reassess3 monthsNon-muscleinvasiveCSCCTLBMBladdersparingThird induction of BCG then reassessInduction of BCGHigh riskEndoscopic controlRecurrenceManagementMuscleinvasiveRadical cystectomyFIGURE 5 Developed care pathway – follow-up. BM, biomarkers; CSC, flexible cystoscopy; CTL, cytology.

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Low and intermediate riskPatients at low risk of recurrence and progressionhave TaG1 TCC or solitary T1G1 TCC. Those atintermediate risk have TaG2 TCC or multifocalT1G1 TCC. Multiplicity at presentation and atumour recurrence at 3 months have consistentlybeen shown to be key practical predictors offuture recurrence, and many urologists in theUK tailor their cystoscopic follow-up of low- andintermediate-risk patients based on these twofactors for these reasons:(a) Patients who have a solitary tumour atdiagnosis and no tumour recurrence at 3months are followed up at 9 months andthen annually for 4 further years. If at theend of this 5-year follow-up period they haveremained tumour free they are discharged.During the follow-up visits these patientsundergo flexible cystoscopy and in somecentres cytology and/or biomarker tests.Although most patients with a tumourrecurrence will receive TURBT, some mayhave a cystodiathermy and biopsy.(b) Patients with multiple tumours at presentationand no recurrence at 3 months or a solitarytumour at presentation with recurrence at3 months need more intense follow-up andare followed up every 3 months for the firstyear and annually if they remain tumourfree until 10 years and are then discharged.During the follow-up visits patients undergocystoscopy and in some centres cytology and/or biomarker tests. Those who present with atumour at the follow-up visit undergo eitherTURBT or cystodiathermy and biopsy. Thesepatients may be considered for a course of sixintravesical instillations of mitomycin C orepirubicin.(c) Patients with multiple tumours at presentationand recurrence at 3 months have the highestrisk of recurrence and are followed up every 3months for the first 2 years and then annuallythereafter. They are usually offered a courseof six intravesical instillations of mitomycinC or epirubicin. Those who present with atumour at the follow-up visit undergo eitherTURBT or cystodiathermy and biopsy.During the follow-up visits patients undergocystoscopy and in some centres cytology and/or biomarker tests. Cystoscopies in the first 2years are usually under general anaesthesiausing a rigid cystoscope.During the follow-up period the status of patientsmay change and they may develop muscle-invasive© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.tumours. It is also possible that patients may dieat any time during follow-up from causes relatedto bladder cancer or from unrelated causes. Theoutlined care pathways in Figures 4 and 5 identifythe areas in which PDD and biomarkers couldbe used in conjunction with the standard tests todiagnose patients with suspected bladder cancerand to follow up those who have been diagnosedwith non-muscle-invasive disease. These patientcare pathways will be used to inform the economicmodel and to establish whether the use of PDD andurine biomarkers reduces recurrence or decreasesprogression at follow-up as a consequence ofaltered treatment.Aim of the reviewThe aim of this review is to assess the clinical andcost-effectiveness of PDD and urine biomarkertests in the detection and follow-up of non-muscleinvasivebladder cancer.This aim is addressed through:• a systematic review of PDD, and urinebiomarker tests (FISH, ImmunoCyt andNMP22) and cytology alone or in combination,in the diagnosis and follow-up of bladdercancer• a structured review of the management ofpatients diagnosed with bladder cancer withassociated costs and outcomes• economic modelling of the cost-effectivenessand cost–utility of alternative approaches in thediagnosis and follow-up of patients with nonmuscle-invasivebladder cancer.The specific objectives of the review are to:• estimate the incremental cost-effectivenessof PDD compared with white light rigidcystoscopy, and biomarkers and urine cytology,in initial diagnosis and follow-up• assess the performance of PDD (1) at the timeof TURBT for newly diagnosed bladder cancerand (2) during follow-up of patients with nonmuscle-invasivedisease• assess the performance of urine biomarkersand cytology in (1) initial diagnosis of bladdercancer and (2) during follow-up of patientswith non-muscle-invasive disease• assess whether PDD reduces recurrence and/or progression of non-muscle-invasive diseasecompared with WLC.19

Definition of the decision problemStructure of the remainderof the reportThe remainder of the report is structured asfollows. Chapter 3 describes the methods forreviewing test performance and effectiveness,Chapter 4 assesses the diagnostic accuracy, andclinical effectiveness in terms of recurrence/progression rates, of PDD compared with WLCand Chapter 5 assesses the test performance ofurine biomarkers (FISH, ImmunoCyt, NMP22) andcytology. Chapter 6 assesses the cost-effectivenessof the tests, Chapter 7 discusses factors relevantto the NHS and other parties, Chapter 8 is adiscussion of the findings and Chapter 9 presentsthe review’s conclusions, including implications forthe NHS and for research.20

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Chapter 3Methods for reviewing testperformance and effectivenessIdentification of studiesStudies were identified by searching electronicdatabases and relevant websites, contactwith experts in the field and the scrutiny ofbibliographies of retrieved papers. Highly sensitiveelectronic searches were conducted to identifyreports of published and ongoing studies on thediagnostic performance of the tests of interest, aswell as the effectiveness of PDD-assisted TURBT.The databases searched were MEDLINE (1966to March Week 3 2008), MEDLINE In-Process(1 April 2008), EMBASE (1980 to Week 132008), BIOSIS (1985 to 27 March 2008), ScienceCitation Index (1970 to 1 April 2008), HealthManagement Information Consortium (HMIC)(March 2008) and the Cochrane ControlledTrials Register (Cochrane Library, Issue 1 2008)as well as current research registers [NationalResearch Register (NRR) Archive (September2007), Current Controlled Trials (CCT) (March2008), ClinicalTrials.gov (March 2008) andWHO International Clinical Trials Registry(March 2008)]. Additional databases searchedTABLE 4 Search resultsDatabaseNumber retrievedPrimary reportsMEDLINE (1966 to March Week 3 2008)/EMBASE (1980 to Week 13 2008)/MEDLINE In- 5373Process (1 April 2008) multifile search (after deduplication in Ovid)Science Citation Index (1970 to 1 April 2008)206 aBIOSIS (1985 to 27 March 2008)60 aCENTRAL (Cochrane Library, Issue 1 2008)2 aHMIC (March 2008)2 aTotal 5643BackgroundCDSR (Cochrane Library, Issue 1 2008) 1DARE (March 2008) 21HTA database (March 2008) 15Medion (March 2008) 0Total 37Total assessed for review 5680Ongoing studiesNRR 33CCT 7ClinicalTrial.gov 1WHO International Clinical Trials Registry 0Total 41a The numbers retrieved from the searches in Science Citation Index, BIOSIS, HMIC and CENTRAL refer to the additionalreports found after excluding those identified from the MEDLINE/EMBASE multifile search.21© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Methods for reviewing test performance and effectiveness22for systematic reviews and other backgroundinformation included the Cochrane Database ofSystematic Reviews (CDSR) (Cochrane Library,Issue 1 2008), Database of Abstracts of Reviewsof Effectiveness (DARE) (March 2008), HealthTechnology Assessment (HTA) database (March2008) and Medion (March 2008). A total of 5680reports were identified (Table 4). In addition, thedetails of 41 potentially relevant ongoing studieswere noted. Reference lists of all included studieswere scanned to identify additional potentiallyrelevant studies. Full details of the search strategiesused and websites consulted are documented inAppendix 1.Inclusion and exclusioncriteriaTypes of studiesThe types of studies considered for reporting testperformance were:• direct (head-to-head) studies in which theindex test and reference standard test wereperformed independently in the same group ofpeople• randomised controlled trials (RCTs) inwhich people were randomised to the indexand comparator test(s) and all received thereference standard test.In the event that there was insufficient evidencefrom direct and randomised studies we consideredundertaking indirect (between-study) comparisonsby meta-analysing studies that compared eachsingle test or combination of tests with thereference standard test, and making comparisonsbetween meta-analyses of the different tests.However, this type of study design is less reliablethan direct studies as differences in diagnosticaccuracy are susceptible to confounding factorsbetween studies. The following types of studieswere considered:• Observational studies, including case series,in which the sample is created by identifyingall people presenting at the point of testing(without any reference to the test results).• Case–control studies in which two groups arecreated, one known to have the target diseaseand one known not to have the target disease,when it is reasonable for all included to gothrough the tests. We excluded case–controlstudies when the control group consisted ofcompletely healthy volunteers, or when thecontrol group consisted of completely healthyvolunteers and people with benign urinaryconditions and it was not possible to calculateresults for the control group minus thehealthy volunteers, such that the spectrum ofdisease and non-disease was unlike that to beencountered in a diagnostic situation.Studies reporting test performance had to reportthe absolute numbers of true positives, falsepositives, false negatives and true negatives, orprovide information allowing their calculation.Studies reporting patient- and/or biopsy-levelanalysis (for PDD) and patient- or specimen-levelanalysis (for biomarkers/cytology) were considered.For assessment of the effectiveness of PDD-assistedTURBT compared with WLC-assisted TURBTin terms of outcomes such as recurrence orprogression we focused on RCTs.Types of participantsThe participants considered were people (1)suspected of having bladder cancer or (2)previously diagnosed with non-muscle-invasivebladder cancer and having follow-up cystoscopicexamination.Index and comparator testsThe following tests and comparators wereconsidered:• PDD (using the photosensitising agents 5-ALA,HAL or hypericin) compared with WLC• urine biomarkers (FISH, ImmunoCyt, NMP22)or cytology either alone or compared with eachother.Studies reporting the test performance ofcombinations of the above tests were alsoconsidered.If the evidence allowed, the following subgroupanalyses were planned:• number of tumours on first cystoscopicexamination• type (e.g. CIS) and grade of tumour (WHO1973 or 2004 classification)• tumour recurrence at the first 3-monthcystoscopic examination following TURBT• diagnostic performance of the different PDDphotosensitising agents

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4• diagnostic performance of the differentcategories of urine biomarkers• for urine biomarkers, whether the urine samplewas voided or obtained by bladder wash.Numerous biomarkers exist that potentially couldhave been included in the review but to make thetask manageable within the given time frame thereview’s steering committee agreed that the reviewshould focus only on those biomarkers regardedas being most clinically relevant. These were seenas being either those approved by the US FDAor the three generally regarded as most useful –FISH, ImmunoCyt and NMP22 – with cytology alsoincluded. It was agreed that the Chairman of theBAUS Section of Oncology should be contactedto canvass the views of the Section’s ExecutiveCommittee on the most relevant biomarkers toconsider. Following this, the Chairman on behalfof the Section suggested that the review shouldassess ImmunoCyt, NMP22, FISH and cytology,and consequently these were the tests that wereincluded in the review.Reference standardThe reference standard considered both forstudies reporting PDD and for studies reportingbiomarkers was histopathological examination ofbiopsied tissue.Types of outcomesThe following outcomes were considered:• for PDD:––test performance in detecting non-muscleinvasivebladder cancer––recurrence rate of bladder tumour overtime following initial resection––progression to muscle-invasive disease• for urine biomarkers/cytology:––test performance in detecting non-muscleinvasivebladder cancer.In any studies reporting the above outcomes,the following outcomes were also considered ifreported:• altered treatment as a result of the tests• acceptability of the tests• interpretability of the tests• quality of life (disease-specific and genericinstruments)• adverse effects.Exclusion criteriaThe following types of report were excluded:• animal models• preclinical and biological studies• reviews, editorials and opinions• case reports• abstracts, as usually insufficient methodologicaldetails are reported to allow critical appraisalof study quality• reports investigating technical aspects of a test• non-English language studies.In addition, studies reporting biomarkers orcytology in which the number of participants in theanalysis was less than 100 were excluded. Studiesreporting cytology that predated the publicationyear of the earliest of the included biomarkerstudies were also excluded.Data extraction strategyOne reviewer screened the titles (and abstracts ifavailable) of all reports identified by the searchstrategy. Full-text copies of all studies deemedto be potentially relevant were obtained andtwo reviewers independently assessed them forinclusion. Any disagreements were resolved byconsensus or arbitration by a third party.Data extraction forms for studies reporting PDDand studies reporting biomarkers/cytology weredeveloped and piloted. One reviewer extracteddetails of study design, participants, index,comparator and reference standard tests andoutcome data, and a second reviewer checked thedata extraction. Any disagreements were resolvedby consensus or arbitration by a third party.Quality assessment strategyTwo reviewers independently assessed the qualityof the included diagnostic studies using QUADAS,a quality assessment tool developed for use insystematic reviews of diagnostic studies. 46 QUADASwas developed through a formal consensus methodand was based on empirical evidence. The originalQUADAS checklist contained 14 questions. TheQUADAS tool was adapted to make it moreapplicable to assessing the quality of studies of testsfor detecting bladder cancer (see Appendix 2 foran example of the modified checklist for PDD).23© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4diagnostic accuracy in any of the three parameters.The SROC curves from the HSROC models wereproduced and are shown on the correspondingSROC plots along with the individual studyestimates. Summary sensitivity, specificity, positiveand negative likelihood ratios and DORs for eachmodel were reported as point estimate and 95%confidence interval (CI).The presentation of test performance in terms ofthe detection of stage and grade of non-muscleinvasivebladder cancer was considered in the twobroad categories of: (1) less aggressive, lower risktumours (pTa, G1, G2) and (2) more aggressive,higher risk tumours (pT1, G3, CIS). The median(range) sensitivity of PDD and WLC across studies,for both patient- and biopsy-based detection oftumours, was reported for each category and alsoseparately for CIS.WLC-assisted TURBT comparedwith PDD-assisted TURBTFor relevant outcomes (e.g. recurrence rate afterWLC-assisted TURBT compared with PDD-assistedTURBT), when appropriate, meta-analysis wasemployed to estimate a summary measure of effect.The dichotomous outcome data were combinedusing the Mantel–Haenszel (RR) method. Forthe estimates of RR, 95% CIs and p-values werecalculated. The results were reported using afixed-effect model in the absence of statisticalheterogeneity. Chi-squared tests and I 2 statisticswere used to explore statistical heterogeneity acrossstudies. Possible reasons for heterogeneity wereexplored using sensitivity analysis. When there wasno obvious reason for heterogeneity, the resultswere reported using random-effects methods.In the event that a quantitative synthesis wasconsidered to be inappropriate or not feasible, weprovided a narrative synthesis of results.25© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Chapter 4Results – photodynamic diagnosisNumber of studiesidentifiedFrom the electronic searches for primary reports,113 records were selected as being possiblyrelevant to the review of PDD. In total, 33 of thesewere non-English language papers and were notconsidered further. The full-text reports of theremaining 80 were obtained and assessed: 44met the inclusion criteria for this review; 25 wereexcluded; and 11 were retained for backgroundinformation. Figure 6 shows a flow diagramoutlining the screening process, with reasons forexclusion of full-text papers.Number and type of studiesincludedAppendix 4 lists the 31 studies, published in 44reports, that were included in the review of testperformance and effectiveness. In total, 27 studies,published in 36 reports, 50–85 met the inclusioncriteria for studies reporting the diagnosticaccuracy of PDD. Four RCTs, published in eightreports, 86–93 met the inclusion criteria for studiescomparing the effectiveness of PDD-assistedTURBT with the effectiveness of WLC-assistedTURBT in terms of outcomes such as recurrence orprogression.Number and type of studiesexcludedA list of the 25 potentially relevant studiesidentified by the search strategy for which full-textpapers were obtained but which subsequently failedto meet the inclusion criteria is given in Appendix5. These studies were excluded because they failedto meet one or more of the inclusion criteriain terms of the type of study, participants, test,reference standard or outcomes reported.Characteristics of theincluded studiesAppendix 6 shows the characteristics of theincluded studies. Table 5 shows summaryinformation for the PDD studies reportingdiagnostic accuracy and Table 6 shows summaryinformation for the RCTs comparing PDD withWLC and reporting recurrence and/or progression.5680 titles/abstracts screened(for both PDD and biomarkers)5600 excluded80 reports selected forfull assessment36 reports excluded:Required outcome not reported: n = 12Required study design not met: n = 10Required reference standard not met: n = 2Comparator not WLC: n = 1Retained for background information: n = 1144 reports of 31 studiesincluded (27 reportingdiagnostic accuracy,4 reporting effectiveness)FIGURE 6 Flow diagram outlining the screening process for the photodynamic diagnosis part of the review.27© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Results – photodynamic diagnosisTABLE 5 Summary of the characteristics of the PDD diagnostic accuracy studiesCharacteristic Number Number of studiesPatientsEnrolled 2949 27Analysed 2807Suspicion of or previously diagnosed BC aSuspicion of BC 946 (41%) 19 (70%)Previously diagnosed BC 1381 (59%)Not reported 481 8 (30%)AgeMedian (range) of means (years) 67 (52–72) 20 (74%)Not reported – 7 (26%)Sex bMen 1647 (76%) 18 (67%)Women 510 (24%)Not reported 656 9 (33%)Agent used5-ALA 2113 (75%) 18 (67%)HAL 464 (17%) 5 (19%)Hypericin 81 (3%) 2 (7%)5-ALA or HAL 149 (5%) 2 (7%)BC, bladder cancer.a Suspicion of or previously diagnosed BC. The totals for this section sum to 2808 rather than 2807 because (1) in thestudy by Fradet and colleagues, 57 of 196 patients included in the analysis, 62 presented with a suspicion of BC, 133had previously diagnosed disease (total of 195) and information was missing for one patient, and (2) in the study byKriegmair and colleagues, 70 29 patients were reported to have presented with suspicion of BC and 77 with previouslydiagnosed BC (total of 106), but only 104 patients were included in the analysis.b Sex. This section sums to 2813 rather than 2807 because the study by Koenig and colleagues 67 reported genderinformation for those enrolled (n = 55) rather than those analysed (n = 49).28The 27 diagnostic studies enrolled 2949participants, with 2807 included in the analysis.In 19 studies 51,53,57–63,65,66,70–72,77,78,80,81,84 involving2327 participants, 946 (41%) presented with asuspicion of bladder cancer and 1381 (59%) hadpreviously diagnosed bladder cancer. In two 72,78of these studies the whole patient population(n = 102) had a suspicion of bladder cancer andin three 51,58,84 the whole population had previouslydiagnosed bladder cancer (n = 117). The remainingeight studies 50,52,54,56,67,73,76,85 did not report thisinformation. In total, 18 50,53,54,56,58,59,61–63,67,70–73,77,78,84,85(67%) of 27 studies used 5-ALA as thephotosensitising agent, five 57,60,65,66,81 (19%) usedHAL, two 52,76 (7%) used hypericin and two 51,80 usedeither 5-ALA or HAL but did report the number ofpatients receiving each agent.Across 20 studies 50,51,53,56,57,59–63,65–67,70,71,76,77,80,81,84providing information on patient age, the median(range) of means was 67 years (52–72 years).In total, 18 studies 50,53,54,57,60–63,65–67,70,71,76,77,80,81,84provided information on the gender of 2157participants, of whom 1647 (76%) were men and510 (24%) were women.Sixteen studies 50,51,53,56,59–63,65,71,76,77,80,81,85 gavedetails of when they took place, with an earlieststart date of February 1994 63 and latest enddate of March 2006. 51 Nine studies took placein Germany, 54,56,58,61,70,71,80,84,85 three in theNetherlands, 59,60,81 two each in Italy 51,53 andSingapore 50,76 and one each in Belgium, 52Switzerland, 63 France, 72 Austria, 73 Poland, 78 SouthKorea 62 and China, 77 and four had multinational

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4TABLE 6 Summary of the characteristics of the RCTs reporting recurrence/progressionCharacteristic Number Number of studiesPatientsEnrolled 709 4Analysed 544Suspicion of or previously diagnosed BCSuspicion of BC 48 1Previously diagnosed BC 74Not reported 422 3Age aPDD groups (years) 68 (68–69) 3WLC groups (years)70 (all three studies)Whole study population (years) 67 1SexMen 396 (73%) 4Women 148 (27%)Agent used5-ALA 544 4Outcomes reportedRecurrence-free survival – 2Residual tumour at first cystoscopy – 4Recurrence of tumour – 2Progression to muscle-invasive disease – 2Length of follow-up8 years 15 years 12 years 110–14 days 1BC, bladder cancer.a Age. Babjuk and colleagues, 86 Denzinger and colleagues 89 and Kriegmair and colleagues 92 provided information on patientage separately for the PDD and WLC groups – the information in the table is the median (range) of means across thethree studies. Daniltchenko and colleagues 88 reported the mean age for the study population overall.settings, taking place in the USA/Canada, 57Germany/the Netherlands, 66 Germany/USA 67 andSwitzerland/Norway/Sweden/Germany. 65The four RCTs reporting recurrence/progressionenrolled 709 participants, of whom 544 wereincluded in the analysis. In the study by Babjukand colleagues, 86 of 128 patients enrolled, six wereexcluded because of no histological evidence ofbladder cancer (n = 2), muscle-invasive bladdercancer (n = 3) and multiple T1G3 tumour© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.with concomitant CIS treated with immediatecystectomy (n = 1). In the study by Daniltchenkoand colleagues, 88 115 patients were randomised,with 13 patients subsequently excluded becauseof muscle-invasive bladder cancer. In the studyby Denzinger and colleagues, 89 301 patientswere randomised to the PDD (n = 151) andWLC (n = 150) arms. A total of 63 patients weresubsequently excluded from the PDD arm becauseof no positive tumour confirmation (n = 38),invasive tumour or indication for cystectomy29

Results – photodynamic diagnosis30(n = 23), or no follow-up examinations (n = 2),and 47 patients were excluded from the WLCarm because no tumour could be found (n = 22),muscle-invasive urothelial carcinoma was diagnosedor cystectomy was indicated (n = 23) or follow-upwas refused after the first resection (n = 2, onewith pTaG1 and one with pT1G2). In the studyby Kriegmair and colleagues, 92 of 165 patientsrandomised, 129 patients had histological proof ofTCC and were considered evaluable.The outcomes reported for the studies includedrecurrence-free survival, 86,89 residual tumourrate at first cystoscopy following TURBT, 86,88,89,92recurrence during follow-up 88,89 and progression tomuscle-invasive disease. 88,89Although the selection criteria for all four studiesallowed the inclusion of patients with either asuspicion of or previously diagnosed bladdercancer, only the study by Babjuk and colleagues 86provided details of these groups. Babjuk andcolleagues reported that 20/60 (33%) of thePDD group and 28/62 (45%) of the WLC grouppresented with a suspicion of bladder cancerwhereas 40/60 (67%) of the PDD group and34/62 (55%) of the WLC group had previouslydiagnosed bladder cancer. The remaining studiesby Daniltchenko and colleagues, 88 Denzingerand colleagues 89 and Kriegmair and colleagues 92involving 422 patients did not provide separatedetails of those with a suspicion of bladder cancerand those with previously diagnosed disease. Allfour studies used 5-ALA as the photosensitisingagent.Three studies 86,89,92 provided information onpatient age separately for the PDD and WLCgroups, with the median (range) of means 68 years(68–69 years) for the PDD groups and 70 years(all three studies) for the WLC groups. The studyby Daniltchenko and colleagues 88 reported themean age for the whole patient population as 67years. All four studies provided information onthe gender of the 544 patients analysed, of whom396 (73%) were men and 148 (27%) were women.There were 197 men in the PDD groups and 199 inthe WLC groups, and there were 67 women in thePDD groups and 81 in the WLC groups. All fourstudies gave details of when they took place, withan earliest start date of 1997 89 and latest end dateof December 2003. 86 One (single centre) study tookplace in Germany, 89 one in the Czech Republic, 86and the remaining two were multicentre, with bothtaking place in Germany/Austria. 88,92 The follow-upperiods for the studies were 8 years for Denzingerand colleagues, 89 5 years for Daniltchenko andcolleagues, 88 2 years for Babjuk and colleagues 86and 10–14 days for Kriegmair and colleagues, 92although Kriegmair and colleagues comparedPDD and WLC with the aim of evaluating residualtumour following TURBT, hence the short followupperiod.Quality of the includedstudiesFigure 7 summarises the quality assessment for thePDD diagnostic studies, and Figure 8 summarisesthe quality assessment for the four RCTs thatcompared PDD with WLC and reported recurrence/progression of disease. The results of the qualityassessment of the individual studies are shown inAppendix 7.The diagnostic studies were assessed using amodified version of the QUADAS tool containing13 questions. In 96% (26/27) of studies thespectrum of patients who received the tests wasconsidered to be representative of those who wouldreceive the test in practice. For this question weconsidered patients to be representative if thepatient population either had a suspicion or ahistory of bladder cancer or contained patientsfrom both groups, or the majority or all of thepatient population presented with either grossor microhaematuria or contained a mixture ofpatients with either indication. In all studies thereference standard (histological assessment ofbiopsied tissue) was considered likely to correctlyclassify bladder cancer, and the time periodbetween PDD and the reference standard wasconsidered to be short enough to be reasonablysure that the patient’s condition had not changedbetween the tests.In all studies partial verification bias was avoided inthat all patients who underwent PDD also receiveda reference standard test. However, in only 55%(15/27) of studies 50–54,57–60,63,65,67,70,72,84 were patientsconsidered to have received the same referencestandard regardless of the PDD test result. Thisquestion was checked ‘yes’ if random biopsieswere taken from normal-appearing areas (i.e.test negative) and ‘no’ if biopsies were taken onlyfrom suspicious looking areas (i.e. test positive).In effect the patients in those studies in whichrandom biopsies of normal-appearing areas weretaken received an enhanced reference standard. Inall studies test review bias was avoided in that thePDD results were interpreted without knowledge

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Spectrum representative?Reference standard correctly classifies condition?Time period between tests short enough?Partial verification bias avoided?Differential verification bias avoided?Incorporation bias avoided?Test review bias avoided?Diagnostic review bias avoided?Clinical review bias avoided?Uninterpretable results reported?Withdrawals explained?Positive result clearly defined?Data on observer variation reported?YesUnclearNo0 2040 60 80 100PercentageFIGURE 7 Summary of quality assessment of PDD diagnostic studies (n = 27).Sequence generation really random?Treatment allocation adequately concealed?Groups similar at baseline?Eligibility criteria specified?Intervention clearly defined?Similar treatment apart from intervention?Follow-up adequate for outcomes of interest?Outcomes assessor blinded?Care provider blinded?Patients blinded?Point estimates/variability measures presented?Drop-out rate likely to cause bias?Intention-to-treat analysis undertaken?Operator experienced in the procedure?YesUnclearNo0 2040 60 80 100PercentageFIGURE 8 Summary of quality assessment of RCTs reporting recurrence/progression (n = 4).of the results of the reference standard test. Weconsidered that this would always be the case, aslesions considered suspicious during PDD arebiopsied during the procedure and it is only laterthat the reference standard results are knownfollowing histological assessment of the biopsiedtissue.In 96% (26/27) of studies, either uninterpretableor intermediate test results were reported orthere were no uninterpretable or intermediatetest results, and withdrawals from the study wereexplained or there were none. The exception tothis was the study by Koenig and colleagues, 67in which 55 patients were included but only49 reported in the analysis. In 81% (22/27) ofstudies 50,52–54,56,58–63,65–67,70–73,76–78,84 a clear definitionof what was considered to be a positive result wasprovided. In 96% (26/27) of studies it was unclearwhether the same clinical data were availablewhen the PDD test results were interpretedas would be available when the test was usedin practice, the exception being the study byEhsan and colleagues, 54 which stated that adetailed review of personal medical history wasconducted for each patient before PDD. In thiscontext clinical data were defined broadly toinclude any information relating to the patient31© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Results – photodynamic diagnosis32such as age, gender, presence and severity ofsymptoms, and other test results. In 59% (16/27)of studies 50,53,54,56,58,60,67,70–73,77,78,81,84,85 it was unclearwhether the reference standard results wereinterpreted without knowledge of the results ofthe PDD test. All of the studies were judged tosuffer from incorporation bias in that PDD wasnot considered to be independent of the referencestandard test as the biopsies used for the referencestandard were obtained via the PDD procedure.None of the studies provided information onobserver variation in interpretation of test results.The four RCTs, 86,88,89,92 comparing PDD with WLC,were assessed using the 14-question checklistadapted from Verhagen and colleagues. 47 In all fourstudies the groups were considered to be similar atbaseline in terms of prognostic factors, eligibilitycriteria for the study were specified, and length offollow-up was considered adequate in relation tothe outcomes of interest reported by the studies.In all four studies it was unclear whether thesequence generation was really random, whetherthe treatment allocation was adequately concealed,whether the outcome assessors, care providersor patients were blinded to the PDD or WLCintervention, or whether the surgeon undertakingthe operation was experienced in performingthe procedure. In the studies by Denzinger andcolleagues 89 and Kriegmair and colleagues 92 thewithdrawal rate was considered likely to causebias. In the studies by Babjuk and colleagues 86and Denzinger and colleagues 89 the groups wereconsidered to have been treated in the same wayapart from the intervention received, whereas inthe remaining two studies 88,92 this was unclear. Inthe studies by Daniltchenko and colleagues 88 andDenzinger and colleagues 89 point estimates andmeasures of variability were presented for theprimary outcome measures. Only the study byKriegmair and colleagues 92 included an intentionto treat analysis.Assessment of diagnosticaccuracyOverviewThis section reports the diagnostic accuracy of PDDcompared with WLC against a reference standardof histological assessment of biopsied tissue for thedetection of non-muscle-invasive bladder cancer.The following levels of analysis are presented:patient, biopsy, stage/grade and photosensitisingagent used. For patient and biopsy levels of analysisfigures are included showing the sensitivity andspecificity of the individual studies, SROC curvesand pooled estimates with 95% CIs for sensitivity,specificity, positive and negative likelihood ratiosand DORs for PDD and WLC. For the stage/gradelevel of analysis the median (range) sensitivityand specificity across studies are presented forPDD and WLC. Appendix 8 shows the studies thatreported sufficient information (true and falsepositives and negatives for both PDD and WLC)to allow their inclusion in the pooled estimates forpatient- and biopsy-level analysis, and also thosestudies comparing PDD with WLC that reportedthe sensitivity of the tests in detecting tumourstage/grade. Individual study results are givenin Appendix 9. The results of studies reportingsensitivity and specificity for PDD but not WLCwere examined to assess whether they differed fromthose of the comparative studies.Patient-level analysisAlthough biopsy-level analysis is useful to validatethe accuracy of the test, patient-level data aremore useful in determining management. Fivestudies 65,66,73,80,81 comparing PDD with WLC andenrolling 386 people, with 370 included in theanalysis, provided sufficient information to allowtheir inclusion in the pooled estimates for patientlevelanalysis. In four studies, 65,66,80,81 of 318 patientsincluded in the analysis, 131 (41%) had symptomssuggestive of bladder cancer and 187 (59%) hada history of non-muscle-invasive bladder cancer.The study by Riedl and colleagues 73 did not reportthis information. Three of the studies 65,66,81 usedHAL as the photosensitising agent and two 73,80 used5-ALA. In two 65,73 of the studies random biopsies ofnormal-appearing areas were taken.Figure 9 shows the sensitivity and specificity ofthe individual studies, SROC curves and pooledestimates for the sensitivity and specificity of PDDand WLC for patient-based detection of bladdercancer. The pooled sensitivity (95% CI) for PDDwas 92% (80% to 100%) compared with 71% (49%to 93%) for WLC, whereas the pooled specificity(95% CI) for PDD was 57% (36% to 79%) comparedwith 72% (47% to 96%) for WLC. The pooledestimates show that PDD had higher sensitivitybut lower specificity than WLC, with the CIs forthe two techniques overlapping. None of the fivestudies comparing PDD with WLC reported testperformance separately for the group of patientsnewly presenting with a suspicion of bladder canceror for the group with a history of non-muscleinvasivedisease. The DOR values (95% CI) were

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Sensitivity and specificity: individual study resultsSROC plots for PDD and WLC: patient levelWitjes 2005 81 20 89 100 20 79PDD WLCStudy IDJichlinski 2003 65Jocham 2005 66Riedl 1999 73Tritschler 2007 80n5214652100Sens965310093Spec43813357n5214652100Sens73337688Spec437410055100Sensitivity1.00.90.80.70.60.50.40.30.20.10.00.0 0.1 0.2Test: / PDD/ WLC0.30.4 0.5 0.61–Specificity0.70.80.91.00.90.80.70.60.50.40.30.20.10.01.0SensitivityPooled analysis of patient level PDD dataPooled analysis of patient level WLC dataNumber of studiesSensitivity % (95% CI)Specificity % (95% CI)Positive likelihood ratio (95% CI)Negative likelihood ratio (95% CI)DOR (95% CI)592 (80 to 100)57 (36 to 79)2.17 (1.16 to 3.19)0.13 (0.01 to 0.32)16.50 (1.00 to 42.23)Number of studiesSensitivity % (95% CI)Specificity % (95% CI)Positive likelihood ratio (95% CI)Negative likelihood ratio (95% CI)DOR (95% CI)571 (49 to 93)72 (47 to 96)2.57 (0.53 to 4.61)0.40 (0.12 to 0.67)6.44 (1.00 to 14.24)FIGURE 9 Patient-level analysis: sensitivity, specificity, SROC curve and pooled estimates.16.50 (1.00 to 42.23) for PDD and 6.44 (1.00 to14.24) for WLC, with higher DORs indicating abetter ability of the test to differentiate betweenthose with and those without bladder cancer. Acrossstudies the median (range) PPVs were 91% (59% to100%) for PDD and 89% (56% to 100%) for WLC,and NPVs were 60% (32% to 100%) for PDD and23% (20% to 87%) for WLC. However, it should benoted that predictive values are affected by diseaseprevalence, which is rarely constant across studies,and therefore these data should be interpreted withcaution.Three studies 72,77,78 enrolling and analysing 153patients reported patient-based detection for PDDonly and were not included in the pooled estimates.All three studies used 5-ALA and, in one, 72 randombiopsies of normal-appearing areas were taken.Across these three studies the median (range)sensitivity and specificity for PDD were 91% (64%to 100%) and 67% (36% to 67%) respectively. Intwo 72,78 of the studies the whole patient populations(n = 102) had a suspicion of bladder cancer withno previous history of the disease. Landry andcolleagues 72 reported a sensitivity of 64% forPDD, compared with 91% reported by Szygulaand colleagues, 78 whereas both studies reported aspecificity of 67%.Studies that reported patient-level analysis butonly for CIS are considered in the section on stage/grade analysis.Biopsy-level analysisA total of 14 studies 50,53,54,56,59–63,65,70,76,81,85 comparingPDD with WLC and enrolling 1751 people, with1746 included in the analysis, provided sufficientinformation to allow their inclusion in the pooledestimates for biopsy-level analysis (number ofbiopsies: 8574 for PDD analysis, 8473 for WLCanalysis). In nine studies, 53,59–63,65,70,81 involving1408 people, 560 (40%) had symptoms suggestiveof bladder cancer and 848 (60%) had a history ofnon-muscle-invasive bladder cancer. The studiesby Cheng and colleagues, 50 Ehsan and colleagues, 54Filbeck and colleagues, 56 Sim and colleagues 76 andZumbraegel and colleagues 85 did not report thisinformation. Ten studies 50,53,54,56,59,61–63,70,85 used5-ALA as the photosensitising agent and three 60,65,81used HAL, while the study by Sim and colleagues 76used hypericin. In eight studies 50,53,54,59,60,63,65,7033© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Results – photodynamic diagnosisrandom biopsies of normal-appearing areas weretaken.Figure 10 shows the sensitivity and specificity ofthe individual studies, SROC curves and pooledestimates for the sensitivity and specificity of PDDand WLC for biopsy-level detection of bladdercancer. In the pooled estimates, PDD had highersensitivity (93%, 95% CI 90% to 96%) than WLC(65%, 95% CI 55% to 74%), whereas WLC hadhigher specificity (81%, 95% CI 73% to 90%) thanPDD (60%, 95% CI 49% to 71%). The pair of CIsfor both sensitivity and specificity did not overlap,providing evidence of a difference in diagnosticperformance between the techniques. Across the 14studies the sensitivity for PDD ranged from 76% 65to 98% 54,62,70 compared with 17 53 to 88% 60 for WLC,and specificity ranged from 32% 85 to 100% 81 forPDD compared with 46 85 to 100% 81 for WLC. Inthe pooled analysis the DOR values (95% CI) were20.29 (9.20 to 31.37) for PDD and 7.76 (3.39 to11.93) for WLC. Across studies the median (range)PPVs were 61% (40% to 100%) for PDD and 70%(38% to 100%) for WLC, and the median (range)NPVs were 92% (20% to 99%) for PDD and 78%(13% to 91%) for WLC.None of the 14 studies comparing PDD with WLCreported biopsy-level detection separately for thegroup of patients newly presenting with a suspicionof bladder cancer or for the group with a history ofnon-muscle-invasive disease.Six studies 58,67,71,73,77,84 involving 428 patientsreported biopsy-level detection for PDD only andwere not included in the pooled estimates. All sixstudies used 5-ALA and in four 58,67,73,84 randombiopsies of normal-appearing areas were taken.Across the six studies the median (range) sensitivityand specificity for PDD were 95% (87% to 98%)and 51% (36% to 67%) respectively. In two 58,84 ofthese studies the whole patient population (n = 68)had a history of non-muscle-invasive bladdercancer. Frimberger and colleagues 58 and Zaak andcolleagues 84 reported sensitivities of 95% and 90%and specificities of 67% and 61%, respectively, forPDD.Studies that reported biopsy-level analysis but onlyfor CIS are included in the section on stage/gradeanalysis.Sensitivity and specificity: individual study resultsPDDWLCSROC plots for PDD and WLC: patient levelStudy IDCheng 2000 50De Dominicis 2001 53Ehsan 2001 54Filbeck 1999 56Grimbergen 2003 59Hendricksen 2006 60Hungerhuber 2007 61Jeon 2001 62Jichlinski 1997 63Jichlinski 2003 65Kriegmair 1996 70Sim 2005 76Witjes 2005 81Zumbraegel 2003 85n175179151347917217463027421542143317928408Sens8987989697949298897698828594Spec65636535495856435779649010032n175179151347917123463027421541443317928408Sens6617606969887659464673627480Spec848858667886869157936998100461.00.90.80.70.60.50.40.30.20.10.00.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9Test: PDD1–Specificity// WLCSensitivity1.00.90.80.70.60.50.40.30.20.10.01.0SensitivityPooled analysis of biopsy level PDD dataNumber of studiesSensitivity % (95% CI)Specificity % (95% CI)Positive likelihood ratio (95% CI)Negative likelihood ratio (95% CI)DOR (95% CI)1493 (90 to 96)60 (49 to 71)2.33 (1.73 to 2.92)0.12 (0.06 to 0.17)20.29 (9.20 to 31.37)Pooled analysis of biopsy level WLC dataNumber of studiesSensitivity % (95% CI)Specificity % (95% CI)Positive likelihood ratio (95% CI)Negative likelihood ratio (95% CI)DOR (95% CI)1465 (55 to 74)81 (73 to 90)3.38 (2.01 to 4.75)0.44 (0.33 to 0.54)7.76 (3.39 to 11.93)34FIGURE 10 Biopsy-level analysis: sensitivity, specificity, SROC curve and pooled estimates.

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Formal comparison of PDD andWLC in patient- and biopsybasedanalysisIn addition to the two HSROC models of thediagnostic accuracy of PDD and WLC individually,two HSROC models were run that simultaneouslymodelled PDD and WLC diagnostic accuracyusing all of the data from the 14 studies. Therewas strong evidence of a difference in diagnosticaccuracy between the tests, with the model thatallowed for a difference in diagnostic accuracyin the three constituent parameters (threshold,accuracy and shape of SROC curve) having asubstantially better Bayesian information criterionthan the simplified diagnostic accuracy model, forboth patient- and biopsy-level analysis (differenceof 1408.0 and 20.7 respectively). These resultsare supported by noting that the intervals for thesummary sensitivity and specificity at biopsy levelfrom the models in which the tests were modelledseparately (Figure 10) did not overlap for eithermeasure. PDD had a greater sensitivity than WLCbut at the cost of a lower specificity. The pointestimates of the patient-level analysis were similarto those from the biopsy-level analysis, althoughthe intervals were substantially wider, as might beexpected because of the smaller number of studiesand observations available for this level of analysis.Stage/grade analysisStudies reporting the sensitivity of PDD comparedwith WLC in the detection of stage and grade oftumour categorised this information in differentways, including pTa, pTaG1, pTaG1–2, pTaG2,pTaG2–3, pTaG3, pTa-T1, G1–2, pT1, pT1G1,pT1G1–2, pT1G2, pT1G3, > pT1, CIS, G3,pT2G2, pT2G3, ≥ pT2, ≥ pT2G3 and pT4G3 (seeAppendix 8). Some studies reported the detectionof stage/grade at the patient level and othersreported this information at biopsy level.For the purposes of this review, the presentation oftest performance in terms of the detection of stageand grade of non-muscle-invasive bladder cancerwas considered in two broad categories:1. less aggressive, lower risk tumours (pTa, G1,G2)2. more aggressive, higher risk tumours (pT1, G3,CIS).Table 7 shows the median (range) sensitivity ofPDD and WLC across studies, for both patientandbiopsy-based detection of tumours, within thebroad categories of less aggressive/lower risk andmore aggressive/higher risk (including CIS), andalso separately for CIS.TABLE 7 Sensitivity of PDD and WLC in detecting stage/grade of tumourPDD sensitivity (%),median (range)WLC sensitivity (%),median (range)Number of patients(biopsies) aNumber ofstudiesLess aggressive/lower riskPatient-based detection 92 (20 to 95) 95 (8 to 100) 266 3Biopsy-based detection 96 (88 to 100) 88 (74 to 100) 1206 (5777) 7More aggressive/higher risk including CISPatient-based detection 89 (6 to 100) 56 (0 to 100) 563 6Biopsy-based detection 99 (54 to 100) 67 (0 to 100) 1756 (7506) 13CISPatient-based detection 83 (41 to 100) 32 (0 to 83) 563 6Biopsy-based detection 86 (54 to 100) 50 (0 to 68) 1756 (7506) 13a The number of biopsies is the overall total reported by the studies. In some studies more biopsies were taken for PDDthan for WLC and in these cases the higher number used for PDD has been used in the table. In the less aggressive/lower risk category, Hendricksen and colleagues 60 reported 217 biopsies for PDD and 123 for WLC and Koenig andcolleagues 67 reported 130 biopsies for PDD and 67 for WLC. Hendricksen and colleagues and Koenig and colleagueswere also included in the more aggressive/higher risk category, as was Jichlinski and colleagues, 65 who reported 421biopsies for PDD and 414 for WLC. The studies by Hendricksen and colleagues, Jichlinski and colleagues and Koenig andcolleagues were also amongst those reporting detection of CIS.35© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Results – photodynamic diagnosisLess aggressive, lower risk tumours (pTa,G1, G2)Nine studies 54,56,60–62,66,67,80,81 involving 1452 patientsreported the sensitivity of PDD compared withWLC for the detection of less aggressive, lowerrisk tumours. The stages/grades reported bythese studies included pTa, 54,62,66,80 pTaG1, 60,61,67pTaG1–2, 56 pTaG2 60,61,67,81 and G1–2. 80 Across threestudies 66,80,81 involving 266 patients reportingpatient-based tumour detection, the median(range) sensitivities of PDD at 92% (20% to 95%)and WLC at 95% (8% to 100%) were broadlysimilar. Across seven studies 54,56,60–62,67,81 involving1206 patients reporting biopsy-based tumourdetection (n = 5777 biopsies overall), the median(range) sensitivity of PDD at 96% (88% to 100%)was higher than that of WLC at 88% (74% to 100%)(Table 7).None of the studies reported the specificity of PDDor WLC in detecting less aggressive, lower risktumours.More aggressive, higher risk tumours(pT1, G3, CIS)Sixteen studies 50,51,53,54,56,57,60–62,65–67,70,80,81,85 involving2155 patients reported the sensitivity of PDDcompared with WLC for the detection of moreaggressive, higher risk tumours. The stages/gradesreported by these studies included pTaG2–3, 53pTaG3, 60,61,67,81 pTa-T1, 50,70 pT1, 54,62,66,80 pT1G1, 61pT1G1–2, 60,61,67 pT1G2 60,61,67 pT1G3, 56,60,61,67,81> pT1, 56 G3 80 and CIS. 50,51,53,54,56,57,60–62,65–67,70,80,81,85Across six studies 51,57,65,66,80,81 involving 563patients reporting patient-based tumourdetection, the median (range) sensitivity ofPDD at 89% (6% to 100%) was much higherthan that of WLC at 56% (0% to 100%). Across13 studies 50,53,54,56,57,60–62,65,67,70,81,85 involving 1756patients reporting biopsy-based tumour detection(n = 7506 biopsies overall), the median (range)sensitivity of PDD at 99% (54% to 100%) was alsomuch higher than that of WLC at 67% (0% to100%) (Table 7).None of the studies reported the specificity ofPDD or WLC in detecting more aggressive, higherrisk tumours, other than for CIS, discussed in thefollowing section.Carcinoma in situAlthough CIS is included in the more aggressive/higher risk category reported above, it may alsobe useful to consider separately the performanceof PDD compared with WLC for the detection ofCIS. The same 16 studies 50,51,53,54,56,57,60–62,65–67,70,80,81,85reporting the sensitivity of PDD compared withWLC for the detection of more aggressive/higherrisk tumours also provided this informationspecifically for CIS.Across six studies 51,57,65,66,80,81 involving 563 patientsreporting patient-based tumour detection, themedian (range) sensitivity of PDD for detectingCIS was 83% (41% to 100%), much higher thanthe sensitivity of 32% (0% to 83%) for WLC. Across13 studies 50,53,54,56,57,60–62,65,67,70,81,85 involving 1756patients reporting biopsy-based tumour detection(n = 7506 biopsies overall), the median (range)sensitivity of PDD was 86% (54% to 100%), alsomuch higher than that of WLC at 50% (0% to 68%)(Table 7).Three studies 51,57,70 reported the specificity ofPDD and WLC in detecting CIS and one study 52reported this information only for PDD (Table 8).The specificity reported for PDD ranged from61% 70 to 99% 52 whereas that for WLC rangedfrom 68% 70 to 97%. 51 Two 51,70 of the three studiescomparing PDD with WLC reported higherspecificity for WLC whereas the third study 57reported similar specificities for both techniques.In the PDD studies HAL was associated with higherTABLE 8 Specificity of PDD and WLC in detecting carcinoma in situSpecificity (%)StudyUnit ofanalysisNumberin studyNumberwithout CIS PDD agent PDD WLCColombo 2007 51 Patient 49 31 5-ALA/HAL 71 97Fradet 2007 57 Patient 196 138 HAL 82 83D’Hallewin 2000 52 Biopsy 281 139 Hypericin 99 NRKriegmair 1996 70 Biopsy 329 323 5-ALA 61 6836NR, not reported.

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4values than 5-ALA, with hypericin associated withthe highest value. However, these results should beinterpreted with caution as they are based on only asmall number of studies.Photosensitising agent usedTable 9 shows the median (range) sensitivityand specificity across studies for the differentphotosensitising agents used, for both patientandbiopsy-level detection of bladder cancer.Four studies using 5-ALA 72,73,77,78 and three usingHAL 65,66,81 reported patient-level detection ofbladder cancer. Across the studies using 5-ALA themedian (range) sensitivity and specificity were 96%(64% to 100%) and 52% (33% to 67%), respectively,compared with 90% (53% to 96%) sensitivity and81% (43% to 100%) specificity for HAL.A total of 15 studies using 5-ALA, 50,53,54,56,58,59,61,63,67,70,71,73,77,84,85three using HAL 60,65,81 and one usinghypericin 76 reported biopsy-level detection ofbladder cancer. Across the studies using 5-ALA themedian (range) sensitivity and specificity were 95%(87% to 98%) and 57% (32% to 67%), respectively,compared with 85% (76% to 94%) sensitivity and80% (58% to 100%) specificity for HAL. The studyby Sim and colleagues 76 reported 82% sensitivityand 91% specificity for hypericin.The results for both patient- and biopsy-baseddetection suggest that 5-ALA may have slightlyhigher sensitivity than HAL, whereas HAL mayhave higher specificity than 5-ALA, but thisshould be interpreted with caution as factors otherthan the photosensitising agent used may havecontributed to the sensitivity and specificity valuesreported by the studies.Four studies reported sensitivity and specificity atboth patient and biopsy level, two using 5-ALA 73,77and two using HAL. 65,81Side effects of photosensitisingagents5-Aminolaevulinic acidA total of 18 studies used 5-ALA as thephotosensitising agent. Seven studies 53,61,63,71–73,78involving 1320 patients reported that no sideeffects were associated with the instillation of5-ALA. Jeon and colleagues, 62 in a study involving62 patients, reported that there were no systemic orserious local side effects following 5-ALA bladderinstillation.Cheng and colleagues, 50 in a study involving 41patients, reported that besides two (5%) patientswho complained of urgency and were unable toretain ALA for more than 2 hours, there wereno clinically significant short-term side effectssuch as urinary tract infections and phototoxicity.At the 1-month follow-up no phototoxicity orother complications were reported. 50 Koenig andcolleagues, 67 in a study involving 49 patients,reported that none showed signs of systemic sideTABLE 9 Sensitivity and specificity according to photosensitising agent usedAgentSensitivity (%),median (range)Specificity (%),median (range)Number of patients(biopsies)Number ofstudiesPatient-based detection5-ALA 96 (64 to 100) 52 (33 to 67) 205 4HAL 90 (53 to 96) 81 (43 to 100) 218 3Hypericin – – – 0Biopsy-based detection5-ALA 95 (87 to 98) 57 (32 to 67) 1949 (8296) 15HAL 85 (76 to 94) 80 (58 to 100) 122 (666) 3Hypericin 82 91 41 (179) 1Two studies included in the table reported only patient- and/or biopsy-based detection of CIS rather than non-muscleinvasivebladder cancer overall. D’Hallewin and colleagues 52 used hypericin and reported biopsy-based detection of CISwhereas Fradet and colleagues 57 used HAL and reported both patient- and biopsy-based detection of CIS.Two studies used either 5-ALA or HAL but did not report the number of patients receiving each agent and are notincluded in the table. 51,8037© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Results – photodynamic diagnosis38effects of PDD such as phototoxicity. One patientreported transient (< 24 hours) dysuria andone patient developed a urinary tract infection,which was treated with antibiotics. 67 Song andcolleagues, 77 in a study involving 51 patients,reported that one cases of acute cystitis wasaccompanied by haemorrhagic lesion attributed tothe instillation procedure (i.e. chemical cystitis).Kriegmair and colleagues, 71 in a study involving104 patients, reported that no serious side effectswere observed during or after 5-ALA instillation.However, following instillation seven patientsreported urgency. After the PDD procedure, moresevere alginuresis symptoms and pollakiuria weredetected in four patients. Significant gram-negativebacteriuria was detected in three patients but thesymptoms improved rapidly with appropriateantibiotics and spasmolytic agents. Phototoxicitywas not detected in any patient. 71Five studies 54,56,58,59,84 did not mention side effects.HexaminolaevulinateFive studies used HAL as the photosensitisingagent. In the studies by Jichlinski and colleagues 65and Witjes and colleagues 81 adverse eventswere reported in 40 of 52 and 4 of 20 patients,respectively, although none was consideredto be related to HAL instillation. Fradet andcolleagues 57 and Jocham and colleagues 66 bothreported that HAL was well tolerated. In the studyby Fradet and colleagues, 57 800 adverse eventswere reported by 240 of the 298 patients in thesafety set, of which 19 (2.4%) were considered tobe related to HAL instillation, none of which wasserious. Twenty patients experienced a total of23 serious adverse events, including one deathdue to an aortic aneurysm, which was unrelatedto HAL instillation. 57 In the study by Jochamand colleagues 66 75 adverse events were reportedby 47 of 162 patients, of which two (2.7%) wereconsidered treatment related, with both occurringin the same patient (urinary retention andmicturition urgency).The study by Hendricksen and colleagues 60 did notmention side effects.5-Aminolaevulinic acid/hexaminolaevulinate not reportedseparatelyTwo studies 51,80 involving 149 patients used 5-ALAor HAL but did not report the number of patientswho received each agent. In the study by Colomboand colleagues, 51 no systemic side effects relatedto the PDD procedure were reported and any localside effects were referred to as negligible. Tritschlerand colleagues 80 did not mention side effects.HypericinTwo studies used hypericin. D’Hallewin andcolleagues, 52 in a study involving 40 patients,reported that there were no significant local orsystemic side effects caused by the instillation ofhypericin. In the study by Sim and colleagues, 76involving 41 patients, there were no reportsof urinary tract infections, contracted bladder,photosensitivity or allergies. One patient developedmicroscopic haematuria from cystitis, whichresolved on conservative management. 76Recurrence/progression ofdiseaseOverviewThis section presents the results of the fourRCTs 86,88,89,92 comparing PDD with WLCand reporting the effectiveness outcomes ofrecurrence-free survival, residual tumour rateat first cystoscopy following TURBT, recurrencerate during follow-up and tumour progression.Random-effects meta-analyses using RR as theeffect measure are presented comparing PDD andWLC in terms of these outcomes.The RCTs enrolled 709 participants, with544 included in the analysis. In the study byDaniltchenko and colleagues 88 the groups wererandomised to WLC or PDD, whereas in thestudies by Babjuk and colleagues, 86 Denzingerand colleagues 89 and Kriegmair and colleagues 92the groups were randomised to WLC or WLC andPDD. The follow-up periods varied from 10–14days for the study by Kriegmair and colleagues, 92which evaluated residual tumour following TURBT,to 2 years for the study by Babjuk and colleagues, 865 years for the study by Daniltchenko andcolleagues 88 and 8 years for the study by Denzingerand colleagues. 89 All four studies used 5-ALA as thephotosensitising agent. Individual study results aregiven in Appendix 9.In the study by Babjuk and colleagues 86 noneof the randomised patients with grade 1 orgrade 2 tumours received adjuvant intravesicaltherapy during the study. All patients with grade3 tumours (six in the PDD group and sevenin the WLC group) received intravesical BCGimmunotherapy, based on a standard 6-weekcourse followed by three, weekly instillations (3-

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4week course) at 3, 6 and 12 months. 86 In the studyby Daniltchenko and colleagues 88 none of therandomised patients received adjuvant intravesicaltherapy throughout the study. In the study byDenzinger and colleagues 89 patients with a solitaryprimary tumour staged pTaG1–G2 did not receiverecurrence prophylaxis. Patients with multifocalinvolvement of the bladder staged pTaG1–G2 orpT1G1–G2 underwent mitomycin therapy, andthose with primary stage pT1G3, CIS or treatmentfailure with mitomycin received BCG therapy, withweekly instillations of 120 mg BCG given for 6weeks. 89 The study by Kriegmair and colleagues 92did not state whether adjuvant intravesical therapywas given, although the primary outcome of thisstudy was to evaluate residual tumour 10–14 daysfollowing TURBT.The four RCTs were reported in eight reports.The study for which Denzinger and colleagues 89 isconsidered the primary report was also reported byFilbeck and colleagues, 91 Burger and colleagues 87and Denzinger and colleagues. 90 The primaryreport gave information on recurrence-freesurvival at 2, 4, 6 and 8 years and also tumourrecurrence throughout this follow-up period,overall and for low-, intermediate- and high-riskgroups, as well as reporting residual tumour rate atsecondary transurethral resection (TUR). Filbeckand colleagues 91 reported residual tumour rate6 weeks after initial resection and recurrencefreesurvival at 12 and 24 months. Burger andcolleagues 87 reported recurrence-free survival,and tumour recurrence and progression at 7.1years, and Denzinger and colleagues 90 reportedrecurrence-free survival and tumour recurrenceand progression for a subgroup of patients whopresented with initial T1 high-grade bladdercancer.The study for which Daniltchenko and colleagues 88is considered the primary report was also reportedby Riedl and colleagues. 93 Daniltchenko andcolleagues 88 reported tumour recurrence andprogression during follow-up whereas Riedl andcolleagues reported residual tumour rate at thecontrol TUR. 93Recurrence-free survivalThe studies by Babjuk and colleagues 86 andDenzinger and colleagues 89 involving a total of 313patients reported recurrence-free survival at 12and 24 months. In a random-effects meta-analysiscomparing PDD and WLC in terms of recurrencefreesurvival, the direction of effect of the pooledestimate at both time points favoured PDD overWLC, although the difference was statisticallysignificant only at 24 months (Figure 11). Therewas evidence of substantial statistical heterogeneityReview:Comparison:Outcome:PDD vs WLC for bladder cancerPDD vs WLCRecurrence-free survivalStudy orsubcategoryPDDn/NWLCn/NRR (random)95% CIWeight%RR (random)95% CI Order01 12 monthsBabjuk 2005 86 40/60 24/62 20.73 1.72 (1.20 to 2.47) 0Denzinger 2007 89 79/88 76/103 79.27 1.22 (1.06 to 1.39) 0Subtotal (95% CI) 148 165 100.00 1.40 (0.96 to 2.03)Total event-free: 119 (PDD), 100 (WLC)Test for heterogeneity: χ 2 = 3.91, df = 1(p = 0.05), I 2 = 74.4%Test for overall effect: z = 1.77 (p = 0.08)02 24 monthsBabjuk 2005 86 24/60 17/62 13.93 1.46 (0.88 to 2.43) 0Denzinger 2007 89 79/88 68/103 86.07 1.36 (1.16 to 1.59) 0Subtotal (95% CI) 148 165 100.00 1.37 (1.18 to 1.59)Total event-free: 103 (PDD), 85 (WLC)Test for heterogeneity: χ 2 = 0.08, df = 1(p = 0.78), I 2 = 0%Test for overall effect: z = 4.13 (p < 0.0001)0.1 0.2 0.5 1 2 5 10Favours WLC Favours PDDFIGURE 11 Recurrence-free survival.39© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Results – photodynamic diagnosisbetween the studies at the 12-month time point(I 2 = 74.4%).Denzinger and colleagues 90 also reported on asubgroup of 46 patients who were diagnosed withT1 high-grade bladder cancer, with recurrence-freesurvival rates of 80% (17/21) in the PDD groupcompared with 52% (13/25) in the WLC group atthe 8-year follow-up.Residual tumour rate atfirst cystoscopy followingtransurethral resectionThe studies by Babjuk and colleagues, 86Daniltchenko and colleagues, 88 Denzinger andcolleagues 89 and Kriegmair and colleagues 92involving a total of 534 patients reported residualtumour rate at first cystoscopy following TUR.The timing of the cystoscopy varied between thestudies, with Kriegmair and colleagues 92 reportingthe residual tumour rate 10–14 days after the initialresection, Denzinger and colleagues 89 and Riedland colleagues 93 reporting it 6 weeks after initialresection, and Babjuk and colleagues 86 assessingthe residual tumour rate 10–15 weeks after TUR.Figure 12 shows a random-effects meta-analysiscomparing PDD with WLC in terms of residualtumour (pTa and pT1) detected at first cystoscopyfollowing the initial TUR. The pooled estimatesshow that PDD resulted in both statisticallyReview:Comparison:Outcome:PDD vs WLC for bladder cancerPDD vs WLCResidual tumour rate at first cystoscopy following TURBTStudy orsubcategoryPDDn/NWLCn/NRR (random)95% CIWeight%RR (random)95% CI Order01 pTaBabjuk 2005 86 2/38 10/37 23.67 0.19 (0.05 to 0.83) 0Daniltchenko 2005 88 7/40 13/39 52.70 0.53 (0.23 to 1.18) 0Denzinger 2007 89 2/66 13/73 23.62 0.17 (0.04 to 0.73) 0Subtotal (95% CI) 144 149 100.00 0.32 (0.15 to 0.70)Total events: 11 (PDD), 36 (WLC)Test for heterogeneity: χ 2 = 2.71, df = 2(p = 0.26), I 2 = 26.1%Test for overall effect: z = 2.85 (p = 0.004)02 pT1Babjuk 2005 86 3/22 13/25 46.93 0.26 (0.09 to 0.80) 0Daniltchenko 2005 88 1/11 7/12 19.65 0.16 (0.02 to 1.07) 0Denzinger 2007 89 2/17 9/25 33.42 0.33 (0.08 to 1.33) 0Subtotal (95% CI) 50 62 100.00 0.26 (0.12 to 0.57)Total events: 6 (PDD), 29 (WLC)Test for heterogeneity: χ 2 = 0.37, df = 2(p = 0.83), I 2 = 0%Test for overall effect: z = 3.34 (p = 0.0008)03 OverallBabjuk 2005 86 5/60 23/62 19.50 0.22 (0.09 to 0.55) 0Daniltchenko 2005 88 8/51 20/51 24.82 0.40 (0.19 to 0.82) 0Denzinger 2007 89 4/83 22/98 16.54 0.21 (0.08 to 0.60) 0Kriegmair 2002 92 25/65 38/64 39.15 0.65 (0.45 to 0.94) 0Subtotal (95% CI) 259 275 100.00 0.37 (0.20 to 0.69)Total events: 42 (PDD), 103 (WLC)Test for heterogeneity: χ 2 = 8.92, df = 3(p = 0.03), I 2 = 66.4%Test for overall effect: z = 3.17 (p = 0.002)0.1 0.2 0.5 1 2 5 10Favours PDD Favours WLC40FIGURE 12 Residual tumour (pTa and pT1) at first cystoscopy following TUR. Notes: 1. In the figure, the numbers of patients shownfor the study by Denzinger and colleagues 89 do not include five each from the PDD and WLC groups who at initial resection had CIS. At6 weeks after initial resection none of the five patients in the PDD group were found to have residual CIS but four of five (80%) in theWLC group were found to have residual CIS. 2. Kriegmair and colleagues 92 reported that in an intention to treat analysis 61.5% (40/65) ofpatients in the PDD group and 40.6% (26/64) of patients in the WLC group were tumour free. For the purposes of the meta-analysis thiswas interpreted as 25/64 patients in the PDD group and 38/64 patients in the WLC group having residual tumour.

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4significantly fewer residual pTa tumours (RR 0.32,95% CI 0.15 to 0.70) and fewer pT1 tumours (RR0.26, 95% CI 0.12 to 0.57), with an overall RRof 0.37 (95% CI 0.20 to 0.69) in favour of PDD(Kriegmair and colleagues 92 reported overall ratesonly).The studies by Babjuk and colleagues 86 andDaniltchenko and colleagues 88 also reportedresidual tumour according to grade, and Figure13 shows a fixed-effect meta-analysis comparingPDD with WLC in terms of the grade of residualtumour detected at first cystoscopy following theinitial TUR. The pooled estimates for G3 were notstatistically significant, whereas those for G1 (RR0.13, 95% CI 0.03 to 0.71), G2 (RR 0.32, 95% CI0.16 to 0.64) and overall (RR 0.31, 95% CI 0.18 to0.53) showed a statistically significant differencein favour of PDD. In the study by Babjuk andcolleagues 86 none of the patients with grade 1 orgrade 2 tumours received adjuvant intravesicaltherapy whereas all those with grade 3 tumoursreceived intravesical BCG immunotherapy. In thestudy by Daniltchenko and colleagues 88 none of thepatients received adjuvant intravesical therapy.Tumour recurrence rate duringfollow-upThe studies by Daniltchenko and colleagues 88 andDenzinger and colleagues 89 involving a total of 293patients reported tumour recurrence rate duringReview:Comparison:Outcome:PDD vs WLC for bladder cancerPDD vs WLCResidual tumour rate by gradeStudy orsubcategoryPDDn/NWLCn/NRR (fixed)95% CIWeight%RR (fixed)95% CI Order01 G1Babjuk 2005 86 1/30 10/33 22.13 0.11 (0.01 to 0.81) 0Daniltchenko 2005 88 0/9 1/7 3.87 0.27 (0.01 to 5.70) 0Subtotal (95% CI) 39 40 26.01 0.13 (0.03 to 0.71)Total events: 1 (PDD), 11 (WLC)Test for heterogeneity: χ 2 = 0.23, df = 1(p = 0.63), I 2 = 0%Test for overall effect: z = 2.36 (p = 0.02)02 G2Babjuk 2005 86 3/24 10/22 24.25 0.28 (0.09 to 0.87) 0Daniltchenko 2005 88 5/35 16/39 35.17 0.35 (0.14 to 0.85) 0Subtotal (95% CI) 59 61 59.42 0.32 (0.16 to 0.64)Total events: 8 (PDD), 26 (WLC)Test for heterogeneity: χ 2 = 0.10, df = 1(p = 0.75), I 2 = 0%Test for overall effect: z = 3.18 (p = 0.001)03 G3Babjuk 2005 86 1/6 3/7 6.44 0.39 (0.05 to 2.83) 0Daniltchenko 2005 88 3/7 3/5 8.13 0.71 (0.23 to 2.18) 0Subtotal (95% CI) 13 12 14.57 0.57 (0.21 to 1.56)Total events: 4 (PDD), 6 (WLC)Test for heterogeneity: χ 2 = 0.30, df = 1(p = 0.58), I 2 = 0%Test for overall effect: z = 1.10 (p = 0.27)Total (95% CI) 111 113 100.00 0.31 (0.18 to 0.53)Total events: 13 (PDD), 43 (WLC)Test for heterogeneity: χ 2 = 3.39, df = 5(p = 0.64), I 2 = 0%Test for overall effect: z = 4.20 (p < 0.0001)0.1 0.2 0.5 1 2 5 10Favours PDD Favours WLCFIGURE 13 Residual tumour (G1, G2 and G3) at first cystoscopy following transurethral resection.41© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Results – photodynamic diagnosisthe follow-up period. The follow-up period forthe study by Daniltchenko and colleagues was 5years 88 whereas that for the study by Denzinger andcolleagues was 8 years. 89Figure 14 shows a random-effects meta-analysiscomparing PDD with WLC in terms of the numberof patients who experienced tumour recurrenceduring the follow-up period. Although thedirection of effect for both studies favoured PDDit was statistically significant only in the study byDenzinger and colleagues, and the pooled estimatedid not show a statistically significant differencebetween PDD and WLC (RR 0.64, 95% CI 0.39 to1.06). 89 There was evidence of substantial statisticalheterogeneity between the studies (I 2 = 71.1%).In the study by Daniltchenko and colleagues 88none of the randomised patients received adjuvantintravesical therapy. In the study by Denzingerand colleagues 89 patients with a solitary primarytumour staged pTaG1–G2 (low-risk group) didnot receive adjuvant intravesical therapy. Patientswith multifocal involvement of the bladder stagedpTaG1–G2 or pT1G1–G2 (intermediate-risk group)underwent mitomycin therapy, and those withprimary stage pT1G3, CIS or treatment failure withmitomycin (high-risk group) received BCG therapy,with weekly instillations of 120 mg BCG given for6 weeks. 89 Table 10 shows the recurrence rates forthe low-, intermediate- and high-risk groups overthe 8-year follow-up in the study by Denzingerand colleagues. 89 Although there were consistentlyfewer recurrences for PDD compared with WLCacross all risk groups, the difference in recurrencerates between PDD and WLC was smaller in theintermediate- and high-risk groups, both of whichreceived adjuvant intravesical therapy, albeit withwide CIs.In the subgroup of 46 patients initially diagnosedwith T1 high-grade bladder cancer, Denzingerand colleagues 89 reported recurrence rates of14% (3/21) in the PDD group compared with 44%(11/25) in the WLC group during the follow-upperiod.Time to recurrenceThe studies by Babjuk and colleagues 86 andDaniltchenko and colleagues 88 reported time torecurrence of bladder tumours. In the study byBabjuk and colleagues 86 this was a median of 17.05months for the PDD group and 8.05 months forthe WLC group. Babjuk and colleagues 86 alsoreported a median time to recurrence in patientswith multiple tumours of 13.54 months for thePDD group and 4.45 months for the WLC group.Daniltchenko and colleagues 88 reported a median(range) time to recurrence of 12 months (2 to 58)for the PDD group and 5 months (2 to 52) for theWLC group.Tumour progression duringfollow-upThe studies by Daniltchenko and colleagues 88 andDenzinger and colleagues 89 also reported tumourprogression during their follow-up periods of 5years and 8 years respectively.Figure 15 shows a fixed-effect meta-analysiscomparing PDD with WLC in terms of the numbersof patients who experienced tumour progressionReview:Comparison:Outcome:PDD vs WLC for bladder cancerPDD vs WLCRecurrence during the follow-up periodStudy orsubcategoryPDDn/NWLCn/NRR(random)95% CIWeight%RR (random)95% CI OrderDaniltchenko 2005 88 30/51 38/51 56.91 0.79 (0.60 to 1.04) 0Denzinger 2007 89 18/88 43/103 43.09 0.49 (0.31 to 0.78) 0Total (95% CI) 139 154 100.00 0.64 (0.39 to 1.06)Total events: 48 (PDD), 81 (WLC)Test for heterogeneity: χ 2 = 3.45, df = 1(p = 0.06), I 2 = 71.1%Test for overall effect: z = 1.72 (p = 0.09)0.1 0.2 0.5 1 2 5 10Favours PDD Favours WLC42FIGURE 14 Tumour recurrence rates during the follow-up period.

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4TABLE 10 Tumour recurrence by risk group in the Denzinger study 89Risk groupIntravesical therapy?Recurrence rate (n/N)PDDWLCLow No 7% (6/88) 19% (20/103)Intermediate Yes 7% (6/88) 13% (13/103)High Yes 7% (6/88) 10% (10/103)during the follow-up period. The direction of effectof the study by Daniltchenko and colleagues 88favoured PDD (four versus nine events) whereas inthe study by Denzinger and colleagues 89 there weretwo cases in each group. The pooled estimate hadwide CIs reflecting the small number of events (RR0.57, 95% CI 0.22 to 1.46).In the subgroup of patients diagnosed withT1 high-grade bladder cancer, Denzinger andcolleagues 90 reported progression to muscleinvasivedisease (≥ T2) of 19% (4/21) in the PDDgroup compared with 12% (3/25) in the WLC groupduring the follow-up period.Summary – assessment ofdiagnostic accuracy andrecurrence/progression ofdiseaseAssessment of diagnosticaccuracyA total of 31 studies, published in 44 reports,met the inclusion criteria for the PDD part of thereview. In total, 27 studies (36 reports) reported thediagnostic accuracy of PDD. As measured by themodified QUADAS checklist, in all studies partialverification bias was avoided (all patients receiveda reference standard test) and test review bias wasavoided (PDD and WLC were interpreted withoutknowledge of the results of the reference standardtest). In 96% (26/27) of studies uninterpretable orintermediate test results were reported or therewere none, and withdrawals from the study wereexplained or there were none. However, all of thestudies were judged to suffer from incorporationbias in that PDD was considered not to beindependent of the reference standard test asbiopsies used in the reference standard test wereobtained via the PDD procedure.In both patient- and biopsy-based detection ofbladder cancer PDD had higher sensitivity butlower specificity than those of WLC. Five studiesinvolving 370 patients reported patient-baseddetection. In the pooled estimates the sensitivityfor PDD was 92% (95% CI 80% to 100%) comparedwith 71% (95% CI 49% to 93%) for WLC, whereasthe specificity for PDD was 57% (95% CI 36% to79%) compared with 72% (95% CI 47% to 96%)for WLC, with the CIs for the two techniquesoverlapping. A total of 14 studies involving 1746Review:Comparison:Outcome:PDD vs WLC for bladder cancerPDD vs WLCProgressionStudy orsubcategoryPDDn/NWLCn/NRR (fixed)95% CIWeight%RR (fixed)95% CI OrderDaniltchenko 2005 88 4/51 9/51 83.00 0.44 (0.15 to 1.35) 0Denzinger 2007 89 2/88 2/103 17.00 1.17 (0.17 to 8.14) 0Total (95% CI) 139 154 100.00 0.57 (0.22 to 1.46)Total events: 6 (PDD), 11 (WLC)Test for heterogeneity: χ 2 = 0.72, df = 1(p = 0.40), I 2 = 0%Test for overall effect: z = 1.18 (p = 0.24)0.1 0.2 0.5 1 2 5 10Favours PDD Favours WLCFIGURE 15 Tumour progression rates during the follow-up period.43© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Results – photodynamic diagnosis44patients reported biopsy-based detection (numberof biopsies: 8574 for PDD analysis, 8473 for WLCanalysis). In the pooled estimates the sensitivity forPDD was 93% (95% CI 90% to 96%) compared with65% (95% CI 55% to 74%) for WLC, whereas thespecificity for PDD was 60% (95% CI 49% to 71%)compared with 81% (95% CI 73% to 90%) for WLC.The pair of CIs for both sensitivity and specificitydid not overlap, providing evidence of a differencein diagnostic performance between the techniques.Studies reporting the sensitivity of PDD comparedwith WLC for detecting stage/grade of bladdercancer categorised this information in differentways. For the purposes of this review the detectionof stage/grade was considered in two broadcategories:1. less aggressive, lower risk tumours (pTa, G1,G2)2. more aggressive, higher risk tumours (pT1, G3,CIS).Across three studies 66,80,81 involving 266 patientsreporting patient-based detection of lower risk, lessaggressive tumours, the median (range) sensitivityof PDD at 92% (20% to 95%) was broadly similarto that of WLC at 95% (8% to 100%). Acrossseven studies 54,56,60–62,67,81 involving 1206 patientsreporting biopsy-based detection (n = 5777 biopsiesoverall), the median (range) sensitivity of PDD wasslightly higher at 96% (88% to 100%) comparedwith 88% (74% to 100%) for WLC. Across sixstudies 51,57,65,66,80,81 involving 563 patients reportingpatient-based detection of more aggressive, higherrisk tumours, the median (range) sensitivityof PDD at 89% (6% to 100%) was higher thanthat of WLC at 56% (0% to 100%). Across 13studies 50,53,54,56,57,60–62,65,67,70,81,85 involving 1756patients reporting biopsy-based detection (n = 7506biopsies overall), the median (range) sensitivity ofPDD at 99% (54% to 100%) was again much higherthan that of WLC at 67% (0% to 100%) (Table 7).These results suggest that PDD is much betterthan WLC in detecting more aggressive, higherrisk tumours. However, the results for patient- andbiopsy-based detection for less aggressive, lowerrisk tumours and patient-based detection formore aggressive, higher risk tumours should beinterpreted with caution as they are based on only asmall number of studies.When CIS was considered separately, acrosssix studies 51,57,65,66,80,81 involving 563 patientsreporting patient-based detection, the median(range) sensitivity of PDD for detecting CISat 83% (41% to 100%) was much higher thanthat of WLC at 32% (0% to 83%). Across 13studies 50,53,54,56,57,60–62,65,67,70,81,85 involving 1756patients reporting biopsy-based detection of CIS(n = 7506 biopsies overall), the median (range)sensitivity of PDD at 86% (54% to 100%) was alsomuch higher than that of WLC at 50% (0% to 68%).The results for patient-based detection should beinterpreted with caution as they are based on onlya small number of studies. However, the mediansensitivity across studies reported for patient-baseddetection of CIS (83%) was similar to that reportedfor biopsy-based detection of CIS (86%). Onlythree studies reported the specificity of PDD andWLC for detecting CIS. Two studies 51,70 reportedhigher specificity for WLC (97% versus 71% and68% versus 61% respectively), whereas the third 57reported similar specificity for both techniques(83% for WLC versus 82% for PDD).Of the studies comparing PDD with WLC that wereincluded in the pooled estimates in the presentreview, two 65,73 of five reporting patient-basedanalysis and eight 50,53,54,59,60,63,65,70 of 14 reportingbiopsy-based analysis undertook random biopsiesof normal-appearing areas. Ten 50,53,54,56,60–62,70,81,85of these 14 studies also reported detection of CISlesions. Table 11 shows, for patient- and biopsylevelanalysis and also for detection of CIS lesions,the sensitivity and specificity for PDD and WLCfor those studies included in the pooled estimatesthat undertook random biopsies compared withthose that did not. There did not appear to be anysystematic pattern to the performance of the testsbased on whether or not random biopsies wereundertaken.Most studies (n = 18) used 5-ALA as thephotosensitising agent, with five using HAL,two hypericin and two either 5-ALA or HAL. Inpatient-based detection of bladder cancer, acrossfour studies using 5-ALA 72,73,77,78 and three usingHAL, 65,66,81 the median (range) sensitivity andspecificity for 5-ALA were 96% (64% to 100%)and 52% (33% to 67%), respectively, comparedwith 90% (53% to 96%) sensitivity and 81%(43% to 100%) specificity for HAL. In biopsybaseddetection of bladder cancer, across 15studies 50,53,54,56,58,59,61,63,67,70,71,73,77,84,85 using 5-ALA, themedian (range) sensitivity and specificity for 5-ALAwere 95% (87% to 98%) and 57% (32% to 67%),respectively, compared with 85% (76% to 94%) and80% (58% to 100%) for HAL. One study, by Simand colleagues, 76 used hypericin, reporting 82%sensitivity and 91% specificity. The results for bothpatient- and biopsy-based detection suggest that

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4TABLE 11 Test performance of studies undertaking/not undertaking random biopsiesPDDWLCNumber ofstudiesMediansensitivity (%)(range)Medianspecificity (%)(range)Mediansensitivity (%)(range)Medianspecificity (%)(range)Patient-level analysisRandom biopsies 2 98 (96 to 100) 38 (33 to 43) 75 (73 to 76) 72 (43 to 100)No randombiopsies3 89 (53 to 93) 81 (57 to 100) 79 (33 to 88) 74 (55 to 100)Biopsy-level analysisRandom biopsies 8 92 (76 to 98) 64 (49 to 79) 63 (17 to 88) 81 (57 to 93)No randombiopsies6 93 (82 to 98) 50 (32 to 100) 72 (61 to 80) 89 (46 to 100)Detection of CISRandom biopsies 5 77 (70 to 100) – 23 (0 to 67) –No randombiopsies5 93 (63 to 100) – 57 (5 to 64) –5-ALA may have slightly higher sensitivity thanHAL, whereas HAL may have higher specificitythan 5-ALA, but this should be interpreted withcaution as factors other than the photosensitisingagent used may have contributed to the sensitivityand specificity values reported by the studies.In total, 20 studies reported side effects.Twelve studies 51–53,61–63,65,71–73,78,81 involving 1543patients reported that there were no side effectsor no serious side effects associated with thephotosensitising agent used (5-ALA, eight studies;HAL, two studies; 5-ALA/HAL not reportedseparately, one study; hypericin, one study). In fourstudies 50,67,71,77 involving 245 patients and using5-ALA, reported side effects associated with theagent included nine patients who complained ofurgency, 50,70 four with alginuresis symptoms andpollakiuria, 70 three with significant gram-negativebacteriuria, 70 one with acute cystitis accompaniedby haemorrhagic lesion, 77 one with transientdysuria 67 and one who developed a urinary tractinfection. 67 Two studies 57,66 involving 460 patientsand using HAL reported 21 non-serious side effectsthat were associated with the agent. One study 76involving 41 patients and using hypericin reportedthat one patient developed microscopic haematuriafrom cystitis.In summary, the evidence suggests that PDD hasclinically important better sensitivity but lowerspecificity than WLC in the detection of bladder© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.cancer and, in terms of stage/grade, has highersensitivity than WLC in the detection of moreaggressive, higher risk tumours (pT1, G3, CIS).Assessment of recurrence/progression of diseaseFour RCTs (eight reports) reporting recurrence/progression enrolled 709 participants, with 544included in the analysis. The follow-up periodsvaried from 10–14 days for the study by Kriegmairand colleagues 92 (although the aim of this study wasto evaluate residual tumour following TURBT) to2 years for the study by Babjuk and colleagues, 865 years for the study by Daniltchenko andcolleagues 88 and 8 years for the study by Denzingerand colleagues. 89 All four studies used 5-ALA as thephotosensitising agent.The study by Daniltchenko and colleagues 88reported that none of the patients receivedadjuvant intravesical therapy. In the study byBabjuk and colleagues 86 only patients with grade3 tumours received intravesical therapy. In thestudy by Denzinger and colleagues 89 patients witha solitary primary tumour staged pTaG1–G2 didnot receive intravesical therapy, whereas those withmultifocal tumours staged pTaG1–G2 or pT1G1–G2 underwent mitomycin therapy and those withprimary stage pT1G3, CIS or treatment failure withmitomycin received BCG therapy. The study by45

Results – photodynamic diagnosisKriegmair and colleagues 92 did not state whetherintravesical therapy was given.In all four studies the PDD and WLC groups weresimilar at baseline in terms of prognostic factors,eligibility criteria for the studies were specified,and length of follow-up was considered adequatein relation to the outcomes of interest reportedby the studies. However, in all four studies it wasunclear whether the sequence generation was reallyrandom or whether the treatment allocation wasadequately concealed.The studies by Babjuk and colleagues 86 andDenzinger and colleagues 89 (involving a total of 313patients) reported recurrence-free survival at 12and 24 months. In a random-effects meta-analysisthe direction of effect of the pooled estimateat both time points favoured PDD over WLC,although the difference was statistically significantonly at 24 months (RR 1.37, 95% CI 1.18 to 1.59).The studies by Babjuk and colleagues, 86Daniltchenko and colleagues, 88 Denzinger andcolleagues 89 and Kriegmair and colleagues 92involving a total of 534 patients reported residualtumour rate at first cystoscopy following TURBT.In a random-effects meta-analysis PDD wasassociated with both statistically significantly fewerresidual pTa tumours (RR 0.32, 95% CI 0.15 to0.70) and pT1 tumours (RR 0.26, 95% CI 0.12 to0.57), with an overall pooled estimate RR of 0.37(95% CI 0.20 to 0.69) in favour of PDD. Babjukand colleagues 86 and Daniltchenko and colleagues 88also reported residual tumour according to grade(G1, G2 and G3). In a fixed-effect meta-analysis thepooled estimates for G1 (RR 0.13, 95% CI 0.03 to0.71) and G2 (RR 0.32, 95% CI 0.16 to 0.64) werestatistically significant in favour of PDD, as was theoverall pooled estimate (RR 0.31, 95% CI 0.18 to0.53).Daniltchenko and colleagues 88 and Denzingerand colleagues, 89 in studies involving a total of293 patients, reported tumour recurrence rateduring the follow-up period (5 years and 8 yearsrespectively). In a random-effects meta-analysis ofthe number of patients who experienced tumourrecurrence, although the direction of effect for bothstudies favoured PDD it was statistically significantonly in the study by Denzinger and colleagues, 89and the direction of effect in the pooled estimatealso favoured PDD but was not statisticallysignificant (RR 0.64, 95% CI 0.39 to 1.06). In thestudy by Denzinger and colleagues 89 the recurrencerates were consistently lower for PDD than for WLCacross all three risk groups. However, the differencein the recurrence rates between PDD and WLC wassmaller in the intermediate-risk [PDD 7% (6/88),WLC 13% (13/103)] and high-risk [PDD 7% (6/88),WLC 10% (10/103)] groups that received adjuvantintravesical therapy than in the low-risk group thatdid not [PDD 7% (6/88), WLC 19% (20/103)].Two studies 86,88 reported time to recurrence, bothfavouring PDD. Babjuk and colleagues 86 reported amedian time to recurrence of 17.05 months for thePDD group and 8.05 months for the WLC group,whereas Daniltchenko and colleagues 88 reported amedian (range) time to recurrence of 12 (2 to 58)months for the PDD group and 5 (2 to 52) monthsfor the WLC group.The studies by Daniltchenko and colleagues 88 andDenzinger and colleagues 89 also reported tumourprogression during their respective 5- and 8-yearfollow-up periods. In a fixed-effect meta-analysisof the number of patients who experiencedtumour progression, the direction of effect of thestudy by Daniltchenko and colleagues favouredPDD whereas that of the study by Denzingerand colleagues favoured WLC, although neitherwas statistically significant. The pooled estimatefavoured PDD but again was not statisticallysignificant (RR 0.57, 95% CI 0.22 to 1.46). 88,89In summary, the evidence suggests that, comparedwith WLC, the use of PDD at TURBT resultsin less residual tumour being found at the firstcystoscopy following TURBT, longer recurrencefreesurvival of patients and a longer time torecurrence following TURBT, and may beassociated with a lower rate of tumour recurrenceover time. However, as these results are based ononly a few studies they should be interpreted withcaution. It should also be borne in mind that theadministration of adjuvant intravesical therapyvaried across the studies. Adjuvant intravesicaltherapy following TURBT is standard practicein the UK and much of Europe and can reducerecurrence by up to 50% in the first 2 years. Thefact that in two studies 86,89 only some patientsreceived intravesical therapy and in one 88 nonedid, while in the fourth study 92 this informationwas not reported, makes it difficult to assess whatthe true added value of PDD might be in reducingrecurrence rates in routine practice.46

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Chapter 5Results – biomarkers and cytologyNumber of studiesidentifiedFrom the electronic searches for primary reports,501 records were selected as being possibly relevantto the review of biomarkers and cytology. In total,133 of these were non-English language papersand were excluded from further assessment. Thefull-text reports of the remaining papers wereobtained and assessed: 83 met the inclusioncriteria for this review; 241 were excluded; and 44were retained for background information. Figure16 shows a flow diagram outlining the screeningprocess, with reasons for exclusion of full-textpapers.Number and type of studiesincludedAppendix 10 lists the 71 studies, published in 83reports, that were included in the review of testperformance.Number and type of studiesexcludedA list of the potentially relevant studies identifiedby the search strategy for which full-text paperswere obtained but which subsequently failed tomeet the inclusion criteria is given in Appendix 11.These studies were excluded because they failed tomeet one or more of the inclusion criteria in termsof the type of study, participants, test(s), referencestandard or outcomes reported.Overview of thebiomarkers/cytologychapterThis chapter contains a section on each of theindividual tests followed by a section on studiesthat directly compared tests and concludes with asummary section. The section on each test containsinformation on the characteristics of the included5680 titles/abstracts screened(for both PDD and biomarkers)5312 excluded368 reports selected forfull assessment285 reports excluded:Fewer than 100 participants: n = 119Required test(s) not reported: n = 79Required study design not met: n = 14Required outcomes not reported: n = 13Criteria for control group not met: n = 10Required reference standard not met: n = 3Cytology studies predating the earliestincluded biomarker study: n = 3Retained for background information:n = 4483 reports of 71 studiesincludedFIGURE 16 Flow diagram outlining the screening process for the biomarkers part of the review.47© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Results – biomarkers and cytologystudies, methodological quality of the studies,results of the pooled estimates for patient-levelanalysis, and also information on specimen-levelanalysis, stage/grade analysis and unevaluabletest results. The methodological quality of thebiomarker and cytology studies was assessed usinga modified version of the QUADAS tool containing14 questions. For patient-level analysis, pooledestimates with 95% CIs for sensitivity, specificity,positive and negative likelihood ratios and DORsare presented. For specimen and stage/gradelevel of analysis the median (range) sensitivityand specificity across studies are presented. If thenumber of specimens reported by a study wasone per patient included in the analysis then thiswas considered as a patient-level analysis. Studiesreporting patient- and specimen-level analysis forCIS are included in the section on stage/gradeanalysis. As described in the previous chapter, forthe purposes of this review, the presentation oftest performance in terms of the detection of stageand grade of non-muscle-invasive bladder cancerwas considered in two broad categories: (1) lessaggressive, lower risk tumours (pTa, G1, G2) and(2) more aggressive, higher risk tumours (pT1, G3,CIS).Appendix 12 shows the characteristics of theincluded biomarker/cytology studies, Appendix 13shows the results of the quality assessment of theindividual studies, Appendix 14 shows the studiesthat reported sufficient information (true and falsepositives and negatives) to allow their inclusionin the pooled estimates for each of the tests forpatient-level analysis, and also those studies thatreported specimen-level analysis and also theTABLE 12 Summary of the characteristics of the FISH studiesCharacteristic Number Number of studiesStudy design aCross-sectional diagnostic study 2704 12Case–control 617 2PatientsEnrolled 3321 14Analysed 2961Suspicion of or previously diagnosed BC a,bSuspicion of BC 1012 (45%) 12 (86%)Previously diagnosed BC 1234 (55%)Not reported 765 2 (14%)AgeMedian (range) of means/medians (years) 70 (63 to 72) 7 (50%)Not reported – 7 (50%)Sex cMen 1073 (71%) 7 (50%)Women 439 (29%)Not reported 1799 7 (50%)48BC, bladder cancer.a In the study design and suspicion of or previously diagnosed BC rows the figures in the number column refer tonumbers of patients.b Suspicion of or previously diagnosed BC. The totals for this section sum to 3011 rather than 3321 because (1) in thestudy by Kipp and colleagues, 99 of 124 participants enrolled, 41 presented with a suspicion of BC, 81 had previouslydiagnosed disease (total of 122) and two had previous cancer of the upper urinary tract and did not fall into eithercategory, and (2) two case–control studies 107,108 contained some participants with benign urological conditions who didnot fall into either category.c Sex. This section sums to 3311 rather than 3321 because the study by Moonen and colleagues 102 reported genderinformation for those analysed (n = 95) rather than those enrolled.

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4sensitivity of the tests in detecting tumour stage/grade, Appendix 15 shows the individual studyresults and Appendix 16 shows the cut-offs used bythe studies reporting FISH that were included inthe pooled estimates.Fluorescence in situhybridisationCharacteristics of the includedstudiesA description of each of the 14 included FISHstudies is given in Appendix 12, which contains thecharacteristics of all of the included biomarker andcytology studies listed alphabetically by surname offirst author. Table 12 shows summary informationfor the 14 FISH studies.Twelve studies, reported in 13 papers, 94–106 werediagnostic cross-sectional studies, of whichtwo 101,104 reported consecutive recruitment, andthe remaining two 107,108 were case–control studies.Two studies were multicentre (21 centres, 108 23centres 103 ).The 14 studies enrolled 3321 participants,with 2961 included in the analysis. In 12studies 94,95,97,99,101–108 reporting this information,1012 (45%) presented with a suspicion of bladdercancer and 1234 (55%) had previously diagnosedbladder cancer. In one 103 of these studies thewhole study population (n = 497) had a suspicionof bladder cancer and in two 102,106 the whole studypopulation had previously diagnosed bladdercancer (n = 355). Two studies 98,100 did not reportthis information.Across seven studies 97,99,101–103,106,107 providinginformation on patient age for the whole studypopulation, the median (range) of means/medians was 70 years (63 to 72 years) (Yoderand colleagues 106 reported median rather thanmean age). Seven studies 94,97,99,102,103,106,107 providedinformation on the gender of 1512 participants, ofwhom 1073 (71%) were men and 439 (29%) werewomen.Seven studies 94,99,102–104,106,108 gave details of whenthey took place, with an earliest start date of1996 104 and latest end date of March 2007. 99 Sevenstudies took place in the USA, 97,99,103–106,108 three inGermany 95,98,107 and one each in the Netherlands 102and Israel, 94 and two had multinational settings,taking place in Austria/Italy 101 and the USA/Belgium. 100© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.Methodological quality of theincluded studiesFigure 17 summarises the quality assessment acrossthe 14 FISH studies. The results for the qualityassessment of the individual studies are shownin Appendix 13. In all studies the spectrum ofpatients who received the tests was consideredto be representative of those who would receivethe test in practice (Q1). For this question weconsidered patients to be representative if thepatient population either had a suspicion or ahistory of bladder cancer or contained patientsfrom both groups, or the majority or all of thepatient population presented with either grossor microhaematuria or contained a mixture ofpatients with either indication. In all studies thereference standard (cystoscopy with histologicalassessment of biopsied tissue) was consideredlikely to correctly classify bladder cancer (Q2). Inall studies partial verification bias was avoided inthat all patients who underwent a FISH test alsoreceived a reference standard test (Q4), differentialverification bias was avoided in that patientsreceived the same reference standard regardless ofthe index test result (Q5) and incorporation biaswas avoided in that the reference standard wasindependent of the index test (Q6). In all studieseither uninterpretable or intermediate test resultswere reported or there were none (Q10), andwithdrawals from the study were explained or therewere none (Q11).In 10 studies (71%) the time period between FISHand the reference standard was considered to beshort enough (1 month or less) to be reasonablysure that the patient’s condition had not changedin the intervening period (Q3). In nine studies(64%) test review bias was avoided in that theFISH results were interpreted without knowledgeof the results of the reference standard test (Q7).However, in nine studies (64%) it was unclearwhether the reference standard results wereinterpreted without knowledge of the results ofthe FISH test (diagnostic review bias, Q8) andin eight studies (57%) it was unclear whether thesame clinical data were available when test resultswere interpreted as would be available when thetest is used in practice (clinical review bias, Q9).In this context clinical data were defined broadlyto include any information relating to the patientsuch as age, gender, presence and severity ofsymptoms, and other test results.In 13 studies (93%) a prespecified cut-off value wasused (Q12); in 10 studies (71%) a clear definition ofwhat was considered to be a positive test result was49

Results – biomarkers and cytologySpectrum representative?Reference standard correctly classifies condition?Time period between tests short enough?Partial verification bias avoided?Differential verification bias avoided?Incorporation bias avoided?Test review bias avoided?Diagnostic review bias avoided?Clinical review bias avoided?Uninterpretable results reported?Withdrawals explained?Prespecified cut-off?Positive result clearly defined?Data on observer variation reported?YesUnclearNo0 2040 60 80 100PercentageFIGURE 17 Summary of quality assessment of FISH studies (n = 14).50provided (Q13); and none of the studies providedinformation on observer variation in interpretationof test results (Q14).Assessment of diagnosticaccuracyPatient-level analysisA total of 12 studies 95,97–101,103–108 enrolling 3101people, with 2535 included in the analysis,provided sufficient information to allow theirinclusion in the pooled estimates for patientlevelanalysis. The cut-offs used by these studiesto define a positive test result were consideredsufficiently similar for all of them to be includedin the pooled estimates (see Appendix 16 for adescription of the cut-offs used by each of theFISH studies). Figure 18 shows the sensitivity andspecificity of the individual FISH studies, pooledestimates and SROC curve for patient-baseddetection of bladder cancer. The pooled sensitivityand specificity (95% CI) were 76% (65% to 84%)and 85% (78% to 92%), respectively, and the DOR(95% CI) value was 18 (3 to 32). Across the 12studies the sensitivity for FISH ranged from 53% 107to 96%, 101 and specificity ranged from 45% 101 to97%. 104 The median (range) PPV across studies was78% (27% to 99%) and the median (range) NPVwas 88% (36% to 97%). However, as previouslymentioned, predictive values are affected by diseaseprevalence, which is rarely constant across studies,and therefore these data should be interpreted withcaution.Most of the included studies in the pooledestimates contained a mixture of patients witha suspicion of bladder cancer and those withpreviously diagnosed bladder cancer, althoughtwo studies 98,100 did not report this information. Inthe study by Sarosdy and colleagues 103 all of theparticipants (n = 497) had a suspicion of bladdercancer (sensitivity 69%, specificity 78%) and inthe study by Yoder and colleagues 106 all of theparticipants (n = 250) had previously diagnosedbladder cancer (sensitivity 64%, specificity 73%).Specimen-level analysisThe study by Moonen and colleagues, 102 enrolling105 participants, all of whom had been previouslydiagnosed with bladder cancer, reported specimenlevelanalysis (n = 103), with sensitivity andspecificity of 39% and 90% respectively.Stage/grade analysisStudies reporting the sensitivity of FISH in thedetection of stage and grade of tumour categorisedthis information in different ways, includingpTa, pTaG1, pTaG1–2, pTaG2, pTaG3, G1, G2,pT1, pT1G2, pT1G3, pT1–4, CIS, G3, pT2,pT2–4, ≥ pT2 and pT4 (see Appendix 14). Allof the studies apart from that by Moonen andcolleagues 102 reported the detection of stage/gradeat the patient level (if the number of specimensreported by a study was one per patient included inthe analysis then this was considered as a patientlevelanalysis).For the purposes of this review the presentation oftest performance in terms of the detection of stage

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Sensitivity and specificity: individual study resultsSROC plotStudy IDFriedrich 2003 95Halling 2000 97Junker2006 98Kipp 2008 99May 2007 107Meiers 2007 100Mian 2003 101Sarosdy 2002 108Sarosdy2006 103Skacel 2003 104Sokolova 2000 105Yoder 2007 106n10315112112416662457392473111179250Sens %678160625393967169858564Spec %899683877490458478979273Sensitivity1.00.90.80.70.60.50.40.30.20.10.00.00.10.20.30.4 0.5 0.61–Specificity0.70.80.91.00.90.80.70.60.50.40.30.20.10.01.0SensitivityPooled analysis of FISH at patient levelNumber of studiesSensitivity % (95% CI)Specificity % (95% CI)Positive likelihood ratioNegative likelihood ratioDOR (95% CI)1276 (65 to 84)85 (78 to 92)5.0 (2.5 to 7.6)0.28 (0.17 to 0.40)17.7 (3.2 to 32.2)FIGURE 18 FISH patient-level analysis: sensitivity, specificity, pooled estimates and SROC curve.and grade of non-muscle-invasive bladder cancerwas considered in two broad categories:1. less aggressive, lower risk tumours (pTa, G1,G2)2. more aggressive, higher risk tumours (pT1, G3,CIS).Table 13 shows the median (range) sensitivityof FISH, for both patient- and specimen-baseddetection of tumours, within the broad categoriesof less aggressive/lower risk and more aggressive/higher risk (including CIS), and also separately forCIS.TABLE 13 Sensitivity of FISH in detecting stage/grade of tumourFISH sensitivity (%),median (range)Number of patients(specimens) aNumber ofstudiesLess aggressive/lower riskPatient-based detection 65 (32 to 100) 2164 10Specimen-based detection 27 (22 to 37) 95 (103) 1More aggressive/higher risk including CISPatient-based detection 95 (50 to 100) 2164 10Specimen-based detection 60 (50 to 67) 95 (103) 1CISPatient-based detection 100 (50 to 100) 1067 8Specimen-based detection NR NR 1NR, not reported.a The numbers of patients and specimens are the totals included in the overall analysis by the studies.51© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Results – biomarkers and cytology52Less aggressive, lower risk tumours(pTa, G1, G2)In total, 10 studies 95,97,99–101,103–105,107,108 involving2164 patients reported the sensitivity of FISHfor the patient-based detection of less aggressive,lower risk tumours. Across these studies the median(range) sensitivity of FISH was 65% (32% to 100%)(Table 13). The study by Moonen and colleagues 102reported specimen-based detection (95 patients,103 specimens), with a median (range) sensitivity of27% (22% to 37%).More aggressive, higher risk tumours(pT1, G3, CIS)In total, 10 studies 95,97,99–101,103–105,107,108 involving2164 patients reported the sensitivity of FISH forthe patient-based detection of more aggressive,higher risk tumours. Across these studies themedian (range) sensitivity of FISH was 95% (50%to 100%) (Table 13). The study by Moonen andcolleagues 102 reported specimen-based detection(95 patients, 103 specimens), with a median (range)sensitivity of 60% (50% to 67%).Carcinoma in situAlthough CIS is included in the more aggressive/higher risk category reported above, it may alsobe useful to consider separately the performanceof biomarkers or cytology for the detection of CIS.Eight studies 95,97,99,101,104,105,107,108 involving 1067patients reported the sensitivity of FISH for thepatient-based detection of CIS. Across these studiesthe median (range) sensitivity of FISH was 100%(50% to 100%) (Table 13).Number of tumoursNone of the included studies reported thesensitivity of FISH in detecting varying numbers oftumours.Size of tumoursNone of the included studies reported thesensitivity of FISH in detecting varying sizes oftumour.Unevaluable testsFive studies 98,101–103,108 reported that 65 of 1059 tests(6.1%) could not be evaluated. The other studiesdid not specifically report this information.ImmunoCytCharacteristics of the includedstudiesA description of each of the 10 includedImmunoCyt studies is given in Appendix 12, whichcontains the characteristics of all of the includedbiomarker and cytology studies listed alphabeticallyby surname of first author. Table 14 shows summaryinformation for the 10 ImmunoCyt studies.All 10 studies, reported in 12 papers, 101,109–119 werediagnostic cross-sectional studies. Six reportedconsecutive recruitment. 101,109,110,112,116,118 Two studieswere multicentre (four centres, 111 19 centres 116 ).The 10 studies enrolled 4199 participants, with atleast 3091 included in the analysis (the study byMian and colleagues 113 enrolled 942 participantsbut did not report the number included in theanalysis). In nine studies 101,109–114,116,118 reportingthis information, 890 participants (27%) presentedwith a suspicion of bladder cancer and 2405 (73%)had previously diagnosed bladder cancer. In oneof these studies 118 the whole patient population(n = 301) had a suspicion of bladder cancer and inthree 110,111,113 the whole population had previouslydiagnosed bladder cancer (n = 1499). One study 119did not report this information.Across six studies 101,109,112–114,116 providinginformation on patient age for the participantgroup as a whole, the median (range) of means was68 years (66 to 73 years). Four studies 112,114,116,118provided information on the gender of 1371participants, of whom 1076 (78%) were men and295 (22%) were women.Six studies 111–114,118,119 gave details of when theytook place, with an earliest start date of November1997 112 and latest end date of July 2007. 118The studies took place in Austria, 112 France, 116Germany, 118 Italy, 113 Sweden, 114 Canada 119 and theUSA, 111 with three having multinational settings, alltaking place in Austria/Italy, 101,109,110 although theydid not state that they were multicentre.Methodological quality of theincluded studiesFigure 19 summarises the quality assessment forthe 10 ImmunoCyt studies. The results for thequality assessment of the individual studies areshown in Appendix 13. In all studies the spectrumof patients who received the tests was consideredto be representative of those who would receive thetest in practice (Q1). In all studies the referencestandard (cystoscopy with histological assessmentof biopsied tissue) was considered likely to correctlyclassify bladder cancer (Q2). In all studies partialverification bias was avoided in that all patientswho underwent an ImmunoCyt test also receiveda reference standard test (Q4), differential

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4TABLE 14 Summary of the characteristics of the ImmunoCyt studiesCharacteristic Number Number of studiesStudy designCross-sectional diagnostic study 4199 10Patients aEnrolled 4199 10Analysed 3091+Suspicion of or previously diagnosed BCSuspicion of BC 890 (27%) 9 (90%)Previously diagnosed BC 2405 (73%)Not reported 904 1 (10%)AgeMedian (range) of means (years) 68 (66 to 73) 6 (60%)Not reported – 4 (40%)Sex bMen 1076 (78%) 4 (40%)Women 295 (22%) 6 (60%)Not reported 2819BC, bladder cancer.a Patients. The number for patients analysed is given as 3091+ because the study by Mian and colleagues 113 enrolled 942participants and reported a specimen-based analysis but did not report the number of participants included in theanalysis.b Sex. This section sums to 4190 rather than 4199 because the study by Schmitz-Drager and colleagues 118 reported genderinformation for 292 of 301 participants enrolled.verification bias was avoided in that patientsreceived the same reference standard regardless ofthe index test result (Q5) and incorporation biaswas avoided in that the reference standard wasindependent of the index test (Q6). In all studieseither uninterpretable or intermediate test resultswere reported or there were none (Q10) and aprespecified cut-off value was used (Q12). In eightstudies (80%) the time period between ImmunoCytand the reference standard was considered to beshort enough (1 month or less) to be reasonablysure that the patient’s condition had not changedin the intervening period (Q3). In nine studies(90%) withdrawals from the study were explainedor there were none (Q11) and a clear definitionof what was considered to be a positive result wasprovided (Q13).In all 10 studies (100%) it was unclear whetherdiagnostic review bias had been avoided (Q8), innine studies (90%) it was unclear whether clinicalreview bias had been avoided (Q9) and in seven© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.studies (70%) it was unclear whether test review biashad been avoided (Q7). One study (10%) providedinformation on observer variation in interpretationof test results (Q14).Assessment of diagnosticaccuracyPatient-level analysisEight studies 101,109,111,112,114,116,118,119 enrolling 3041participants, with 2896 included in the analysis,provided sufficient information to allow theirinclusion in the pooled estimates for patientlevelanalysis. The ‘common’ cut-off used by all ofthese studies to define a positive test result was atleast one green or one red fluorescent cell. Figure20 shows the sensitivity and specificity of theindividual studies, pooled estimates and SROCcurve for ImmunoCyt patient-based detectionof bladder cancer. The pooled sensitivity andspecificity (95% CI) were 84% (77% to 91%) and75% (68% to 83%), respectively, and the DOR value53

Results – biomarkers and cytologySpectrum representative?Reference standard correctly classifies condition?Time period between tests short enough?Partial verification bias avoided?Differential verification bias avoided?Incorporation bias avoided?Test review bias avoided?Diagnostic review bias avoided?Clinical review bias avoided?Uninterpretable results reported?Withdrawals explained?Prespecified cut-off?Positive result clearly defined?Data on observer variation reported?0 2040 60 80 100PercentageYesUnclearNoFIGURE 19 Summary of quality assessment of ImmunoCyt studies (n = 10).(95% CI) was 16 (6 to 26). Across the studies thesensitivity for ImmunoCyt ranged from 73% 116to 100%, 114 and specificity ranged from 62% 119 to88%. 118 The median (range) PPV across studies was54% (26% to 70%) and the median (range) NPVwas 93% (86% to 100%).Most of the included studies in the pooledestimates contained a mixture of patients witha suspicion of bladder cancer and those withpreviously diagnosed bladder cancer, althoughone study 119 did not report this information. In thestudy by Schmitz-Drager and colleagues 118 all of theparticipants (n = 280) had a suspicion of bladdercancer (sensitivity 85%, specificity 88%) and inthe study by Messing and colleagues 111 all of theparticipants (n = 326) had previously diagnosedbladder cancer (sensitivity 81%, specificity 75%).Specimen-level analysisTwo studies 110,113 enrolling 1158 participants, all ofwhom had been previously diagnosed with bladdercancer, reported specimen-level analysis (n = 2220specimens). Across the two studies the median(range) sensitivity and specificity were 78% (71% to85%) and 76% (73% to 78%) respectively.Sensitivity and specificity: individual study resultsSROC plotStudy IDLodde 2003 109Messing 2005 111Mian 1999 112Mian 2003 101Olsson 2001 114Piaton 2003 116Schmitz-Drager 2008 118Tetu 2005 119n225326249181114651280870Sens %87818686100738574Spec %6775797169828862Pooled analysis of ImmunoCyt at patient levelNumber of studies8Sensitivity % (95% CI)84 (77 to 91)Specificity % (95% CI)75 (68 to 83)Positive likelihood ratio3.4 (2.3 to 4.5)Negative likelihood ratio0.22 (0.12 to 0.31)DOR (95% CI)15.7 (5.5 to 25.9)Sensitivity1.00.90.80.70.60.50.40.30.20.10.00.00.10.20.30.4 0.5 0.61–Specificity0.70.80.91.00.90.80.70.60.50.40.30.20.10.01.0Sensitivity54FIGURE 20 ImmunoCyt patient-level analysis: sensitivity, specificity, pooled estimates and SROC curve.

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Stage/grade analysisStudies reporting the sensitivity of ImmunoCytin the detection of stage and grade of tumourcategorised this information in different ways,including pTa, pTaG1–2, pTa pT1G3, pTa+CIS,G1, G2, pT1, pT1G1–2, CIS, G3, pT2 and ≥ pT2(see Appendix 14). All of the studies providingthis information, apart from that by Mian andcolleagues, 113 reported the detection of stage/gradeat the patient level (if the number of specimensreported by a study was one per patient included inthe analysis then this was considered as a patientbasedanalysis).Table 15 shows the median (range) sensitivity ofImmunoCyt, for both patient- and specimen-baseddetection of tumours, within the broad categoriesof less aggressive/lower risk and more aggressive/higher risk (including CIS), and also separately forCIS.Less aggressive, lower risk tumours(pTa, G1, G2)Six studies 101,109,111,112,116,119 involving 2502 patientsreported the sensitivity of ImmunoCyt for thepatient-based detection of less aggressive, lowerrisk tumours. Across these studies the median(range) sensitivity of ImmunoCyt was 81% (55%to 90%) (Table 15). The study by Mian andcolleagues 113 reported specimen-based detection(942 participants enrolled, 1886 specimens), with amedian (range) sensitivity of 82% (79 to 84%).More aggressive, higher risk tumours(pT1, G3, CIS)Six studies 101,109,111,112,116,119 involving 2502 patientsreported the sensitivity of ImmunoCyt for thepatient-based detection of more aggressive, higherrisk tumours. Across these studies the median(range) sensitivity of ImmunoCyt was 90% (67%to 100%) (Table 15). The study by Mian andcolleagues 113 reported specimen-based detection(942 participants enrolled, 1886 specimens), with amedian (range) sensitivity of 91% (84% to 100%).Carcinoma in situSix studies 101,109,111,112,116,119 involving 2502 patientsreported the sensitivity of ImmunoCyt for thepatient-based detection of CIS. Across these studiesthe median (range) sensitivity of ImmunoCyt was100% (67% to 100%). The study by Mian andcolleagues, 113 with specimen as the unit of analysis,reported 100% sensitivity for detecting CIS (Table15).Number of tumoursNone of the included studies reported thesensitivity of ImmunoCyt in detecting varyingnumbers of tumours.Size of tumoursMessing and colleagues, 111 in a study involving326 patients, reported ImmunoCyt sensitivities of71%, 84% and 60% in detecting tumours of < 1 cm,1–3 cm and > 3 cm respectively.TABLE 15 Sensitivity of ImmunoCyt in detecting stage/grade of tumourImmunoCyt sensitivity(%), median (range)Number of patients(specimens) aNumber of studiesLess aggressive/lower riskPatient-based detection 81 (55 to 90) 2502 6Specimen-based detection b 82 (79 to 84) 942 (1886) 1More aggressive/higher risk including CISPatient-based detection 90 (67 to 100) 2502 6Specimen-based detection b 91 (84 to 100) 942 (1886) 1CISPatient-based detection 100 (67 to 100) 2502 6Specimen-based detection b 100 942 (1886) 1a The numbers of patients and specimens are the totals included in the overall analysis by the studies.b Specimen-based detection. In the study by Mian and colleagues 113 942 participants were enrolled but it was unclear howmany were included in the analysis.55© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Results – biomarkers and cytology56Unevaluable testsAll 10 studies 101,109–114,116,118,119 provided informationon unevaluable tests. Overall, 279 of 5292 tests(5%) could not be evaluated. Across studies, themedian (range) percentage of tests that wereunevaluable was 5% (1% to 10%).Observer variationMessing and colleagues 111 reported that after1 day of training pathologists were able to passan interobserver training test, achieving 100%concordance on five slides. At one participatinglaboratory 40% of cases were reviewed bytwo observers independently. There was 90%agreement between observers with the finaldiagnosis of disputed cases agreed on by the twopathologists who reviewed these cases together. 111NMP22Characteristics of the includedstudiesA description of each of the 41 included NMP22studies is given in Appendix 12, which contains thecharacteristics of all of the included biomarker andcytology studies listed alphabetically by surname offirst author. Table 16 shows summary informationfor the 41 NMP22 studies.Thirty-one studies, reported in 37papers, 45,80,95,120–153 were diagnostic cross-sectionalstudies. Three 45,126,127 reported consecutiverecruitment. A total of 10 studies, reported in 11papers, 154–164 were case–control studies. Four studieswere multicentre (23 centres, 126 23 centres, 127 13centres, 135 three centres 149 ).The 41 studies enrolled 13,885 participants,with 13,490 included in the analysis. Fivestudies 80,126,127,131,150 involving 2426 participantsused the NMP22 BladderChek point of care test. In33 studies 45,80,95,122,123,125–132,134–142,144,147–151,153,158,159,162,1644478 participants (41%) presented with a suspicionof bladder cancer and 6536 (59%) had previouslydiagnosed bladder cancer. In five 126,135,141,153,162 ofthese studies the whole patient population analysed(n = 2202) had a suspicion of bladder cancer andin 10 123,127,129–131,138,142,144,147,149 the whole populationanalysed had previously diagnosed bladder cancer(n = 4799). Eight studies 120,121,154–157,161,163 did notreport this information.Across 24 studies 80,123,125–129,131,134,136,138,140,141,144,147,149–151,153,154,156,158,159,162providing information onpatient age for the whole study population, themedian (range) of means was 66 years (53 to 71years). A total of 29 studies 80,121–123,125–127,129–131,135–142,144,147,149–151,153,156,158,159,162,163providedinformation on the gender of 10,804 participants,of whom 7818 (72%) were men and 2986 (28%)were women.In total, 16 studies 80,121,122,126,127,135,136,139,150,151,153,155,158,159,162,164gave details of when they took place,with an earliest start date of August 1995 135 andlatest end date of April 2006. 150 Nine studies tookplace in the USA, 126,127,129,139,148,149,153,158,159 fourin Italy 122,123,144,154 and Spain, 128,142,161,162 three inAustria, 134,147,151 Germany 80,95,132 and Japan, 135,136,164two in the UK, 45,141 Turkey 137,163 and India 131,150and one in Greece, 125 Poland, 130 Switzerland, 121Sweden, 155 the Netherlands, 138 South Korea 157 andChina, 156 and two had multinational settings, takingplace in Germany/USA 140 and Saudi Arabia/USA. 120Methodological quality of theincluded studiesFigure 21 summarises the quality assessment forthe 41 NMP22 studies. The results for the qualityassessment of the individual studies are shown inAppendix 13. In all studies the reference standard(cystoscopy with histological assessment of biopsiedtissue) was considered likely to correctly classifybladder cancer (Q2) and withdrawals from thestudy were explained or there were none (Q11).In 40 studies (98%) the spectrum of patientswho received the tests was considered to berepresentative of those who would receive the testin practice (Q1) and incorporation bias was avoided(Q6). In 39 studies (95%) partial verification biaswas avoided (Q4), intermediate test results werereported or there were none (Q10) and a cleardefinition of what was considered to be a positiveresult was provided (Q13).In 36 studies (88%) differential verification bias wasavoided (Q5), in 32 studies (78%) the time periodbetween NMP22 and the reference standard wasconsidered to be short enough (1 month or less) tobe reasonably sure that the patient’s condition hadnot changed in the intervening period (Q3) andin 24 studies (58%) a prespecified cut-off value wasused (Q12).However, in 39 studies (95%) it was unclear whetherclinical review bias had been avoided (Q9), in 29studies (71%) it was unclear whether test reviewbias had been avoided (Q7) and in 27 studies (66%)it was unclear whether diagnostic review bias hadbeen avoided (Q8). A total of 40 studies (98%) did

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4TABLE 16 Summary of the characteristics of the NMP22 studiesCharacteristic Number Number of studiesStudy designCross-sectional diagnostic study 11,236 31Case–control 2649 10PatientsEnrolled 13,885 41Analysed 13,490Suspicion of or previously diagnosed BC aSuspicion of BC 4478 (41%) 33 (80%)Previously diagnosed BC 6536 (59%)Not reported 1812 8 (20%)AgeMedian (range) of means (years) 66 (53 to 71) 24 (59%)Not reported – 17 (41%)Sex bMen 7818 (72%) 29 (71%)Women 2986 (28%) 12 (29%)Not reported 2858BC, bladder cancer.a Suspicion of or previously diagnosed BC. This section sums to 12,826 rather than 13,885 because Giannopoulos andcolleagues 125 reported this information for those analysed (n = 213) rather than those enrolled (n = 234), Lahme andcolleagues 132 reported it for 84 of 169 participants enrolled, Oge and colleagues 137 reported it for those analysed (n = 76)rather than enrolled (n = 114), Ramakumar and colleagues 159 reported it for 57 of 196 participants enrolled, Sanchez-Carbayo and colleagues 162 reported it for 112 of 187 participants enrolled, Shariat and colleagues 147 reported it for thoseanalysed (n = 2871) rather than those enrolled (n = 2951) and Takeuchi and colleagues 164 reported this information for 48of 669 participants enrolled.b Sex. This section sums to 13,662 rather than 13,885 because Chang and colleagues 156 reported this information for 331of 399 participants enrolled, Sanchez-Carbayo and colleagues 162 reported it for 112 of 187 participants enrolled andShariat and colleagues 147 reported it for those analysed (n = 2871) rather than those enrolled (n = 2951).not report information on observer variation ininterpretation of test results (Q14).Assessment of diagnosticaccuracyPatient-level analysisA total of 28 studies 45,80,95,121–123,126–128,130–132,134,137,139–142,144,147,148,150,151,153,159,160,162,163enrolling 10,565participants, with 10,119 included in the analysis,provided sufficient information to allow theirinclusion in the pooled estimates for patientlevelanalysis, using a ‘common’ cut-off of 10 U/ml to define a positive test result. Figure 22 showsthe sensitivity and specificity of the individualstudies, pooled estimates and SROC curve forNMP22 patient-based detection of bladder cancer.© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.The pooled sensitivity and specificity (95% CI)were 68% (62% to 74%) and 79% (74% to 84%),respectively, and the DOR value (95% CI) was8 (5 to 11). Across the 28 studies the sensitivityfor NMP22 ranged from 33% 45 to 100%, 153 andspecificity ranged from 40% 80 to 93%. 142 Themedian (range) PPV across studies was 52% (13% to94%) and the median (range) NPV was 82% (44%to 100%).Most of the included studies in the pooledestimates contained a mixture of patientswith a suspicion of bladder cancer and thosewith previously diagnosed bladder cancer,although three studies 121,160,163 did not reportthis information. In four studies 126,141,153,162 all ofthe participants (n = 1893) had a suspicion of57

Results – biomarkers and cytologySpectrum representative?Reference standard correctly classifies condition?Time period between tests short enough?Partial verification bias avoided?Differential verification bias avoided?Incorporation bias avoided?Test review bias avoided?Diagnostic review bias avoided?Clinical review bias avoided?Uninterpretable results reported?Withdrawals explained?Prespecified cut-off?Positive result clearly defined?Data on observer variation reported?YesUnclearNo0 2040 60 80 100PercentageFIGURE 21 Summary of quality assessment of NMP22 studies (n = 41).58bladder cancer [median (range) sensitivity andspecificity across studies 71% (56% to 100%)and 86% (80% to 87%) respectively]. In sevenstudies 123,127,130,131,142,144,147 all of the participants(n = 4284) had previously diagnosed bladdercancer [median (range) sensitivity and specificityacross studies 69% (50% to 85%) and 81% (46% to93%) respectively].NMP22 BladderChek point of care testFive studies 80,126,127,131,150 involving 2426 participantsused the NMP22 BladderChek point of care test.Across these studies, using a cut-off of 10 U/ml fora positive test result, the median (range) sensitivityand specificity for patient-based detection ofbladder cancer were 65% (50% to 85%) and 81%(40% to 87%), respectively, compared with 68%(95% CI 62% to 74%) sensitivity and 79% (95% CI74% to 84%) specificity for the 28 studies includedin the pooled estimates. (The five studies using theNMP22 BladderChek test were also included inthe pooled estimates.) In the study by Grossmanand colleagues 126 all of the participants (n = 1331)had a suspicion of bladder cancer (sensitivity 56%,specificity 86%). In the studies by Grossman andcolleagues 127 and Kumar and colleagues 131 all ofthe participants (n = 799) had previously diagnosedbladder cancer [median (range) sensitivity andspecificity across studies 68% (50% to 85%) and83% (78% to 87%) respectively].Specimen-level analysisThree studies enrolling 655 participants reportedspecimen-level analysis (n = 705 specimens forOosterhuis 2002 138 and Stampfer 1998; 149 Bhuiyan2003 120 did not report numbers) using a cut-offof 10 U/ml for a positive test result. Across thethree studies the median (range) sensitivity andspecificity were 49% (25% to 50%) and 92% (68% to94%) respectively.Stage/grade analysisStudies reporting the sensitivity of NMP22 in thedetection of stage and grade of tumour categorisedthis information in different ways, including pTa,pTaG1, pTaG1–2, PTaG2, pTa pT1, pTa pT1 CIS,pTa+CIS, pTaG3–pT1, G1, G2, G1–2, G1 G3,pT1, pT1G2, CIS, G3, pT2, pT2 pT2a, pT2G2,pT2–3, pT2–4, ≥ pT2, pT3, pT3a 3b and pT4 (seeAppendix 14). Almost all of the studies providingthis information and using a cut-off of 10 U/mlfor a positive test result reported the detection ofstage/grade at the patient level (if the number ofspecimens reported by a study was one per patientincluded in the analysis then this was consideredas a patient-based analysis); the exception wasthose studies by Oosterhuis and colleagues 138 andStampfer and colleagues. 149Table 17 shows the median (range) sensitivity ofImmunoCyt, for both patient- and specimen-baseddetection of tumours, within the broad categoriesof less aggressive/lower risk and more aggressive/higher risk (including CIS), and also separately forCIS.Less aggressive, lower risk tumours (pTa,G1, G2)A total of 18 studies 45,95,121,123,126–128,131,132,134,137,141,142,144,150,151,159,162involving 4685 patients reported

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Sensitivity and specificity: individual study resultsStudy IDCasella 2000 121Casetta 2000 122Chahal 2001b 45Del Nero 1999 123Friedrich 2003 95Grossman 2005 126Grossman 2006 127Gutierrez Banos 2001 128Kowalska 2005 130Kumar 2006 131Lahme 2001 132Mian 2000 134Oge 2001 137Ponsky 2001 139n23510221110510313316681509813110924076608Sens %5264338370565076538563567488Spec %8463928765868791467861796984Study IDPoulakis 2001 140Ramakumar 1999 159Saad 2002 141Sanchez-Carbayo 1999 160Sanchez-Carbayo 2001a 142Sanchez-Carbayo 2001b 162Serretta 2000 144Shariat 2006 147Sharma 1999 148Sozen 1999 163Talwar 2007 150Tritschler 2007 80Wiener 1998 151Zippe 1999 153n7391961201872321121792871199140196100291330Sens %79538181696175576773676548100Spec %7060879193805581868181406986SROC plotNumber of studiesSensitivity % (95% CI)Specificity % (95% CI)Positive likelihood ratioNegative likelihood ratioDOR (95% CI)Pooled analysis2868 (62 to 74)79 (74 to 84)3.2 (2.4 to 4.0)0.41 (0.33 to 0.49)7.8 (4.5 to 11.1)Sensitivity1.00.90.80.70.60.50.40.30.20.10.00.00.10.20.30.4 0.5 0.61–Specificity0.70.80.91.00.90.80.70.60.50.40.30.20.10.01.0SensitivityFIGURE 22 NMP22 patient-level analysis: sensitivity, specificity, pooled estimates and SROC curve.TABLE 17 Sensitivity of NMP22 in detecting stage/grade of tumourNMP22 sensitivity (%),median (range)Number of patients(specimens) aNumber of studiesLess aggressive/lower riskPatient-based detection 50 (0 to 86) 4685 18Specimen-based detection 33 191 (431) 1More aggressive/higher risk including CISPatient-based detection 83 (0 to 100) 7556 19Specimen-based detection 82 (25 to 100) 191 (431) 1CISPatient-based detection 83 (0 to 100) 3453 11Specimen-based detection 25 191 (431) 1a The numbers of patients and specimens are the totals included in the overall analysis by the studies.59© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Results – biomarkers and cytology60the sensitivity of NMP22 for the patient-baseddetection of less aggressive, lower risk tumours.Across these studies the median (range) sensitivityof NMP22 was 50% (0% to 86%) (Table 17). Thestudy by Oosterhuis and colleagues 138 reported asensitivity of 33% for specimen-based detection(191 participants, 431 specimens).More aggressive, higher risk tumours(pT1, G3, CIS)A total of 19studies 45,95,121,123,126–128,131,132,134,137,141,142,144,147,150,151,159,162involving 7556 patients reported the sensitivityof NMP22 for patient-based detection of moreaggressive, higher risk tumours. Across thesestudies the median (range) sensitivity of NMP22was 83% (0% to 100%) (Table 17). In the study byOosterhuis and colleagues 138 (191 participants,431 specimens), the median (range) sensitivity forspecimen-based detection was 82% (25% to 100%).Carcinoma in situA total of 11 studies 95,126,127,134,137,141,142,144,150,159,162involving 3453 patients reported the sensitivityof NMP22 for the patient-based detection of CIS.Across these studies the median (range) sensitivityof NMP22 was 83% (0% to 100%). Oosterhuis andcolleagues 138 (191 participants, 431 specimens)reported a sensitivity of 25% for specimen-baseddetection of CIS.Number of tumoursThree studies reported the sensitivity of NMP22in detecting bladder cancer in patients withvarying numbers of tumours, although none ofthe studies used a cut-off of 10 U/ml. Poulakis andcolleagues 140 in a study involving 739 patientsreported NMP22 (cut-off ≥ 8.25 U/ml) sensitivitiesof 79%, 90% and 97% in patients with one, two tothree, and more than three tumours respectively.Takeuchi and colleagues 164 in a study involving669 patients reported NMP22 (cut-off ≥ 12 U/ml)sensitivities of 44%, 60% and 91% in patientswith one, two to four, and five or more tumoursrespectively. Sanchez-Carbayo and colleagues 161 ina study involving 187 patients reported NMP22(cut-off ≥ 14.6 U/ml) sensitivities of 72% and 75%in patients with single and multiple tumoursrespectively.Size of tumoursThree studies reported the sensitivity of NMP22in detecting bladder cancer in patients withvarying sizes of tumours, although again none ofthe studies used a cut-off of 10 U/ml. Boman andcolleagues 155 in a study involving 250 patientsreported NMP22 (cut-off ≥ 4 U/ml) sensitivities of65%, 54%, 73% and 89% in detecting new tumoursof ≤ 10 mm, 11–20 mm, 21–30 mm and > 30 mm,respectively, and 41%, 67% and 60% in detectingrecurrent tumours of ≤ 10 mm, 11–20 mm and> 21 mm respectively. Takeuchi and colleagues 164in a study involving 669 patients reported NMP22(cut-off ≥ 12 U/ml) sensitivities of 32%, 65% and92% in detecting tumours < 10 mm, 10–30 mmand > 30 mm respectively. Sanchez-Carbayo andcolleagues 161 in a study involving 187 patientsreported NMP22 (cut-off > 14.6 U/ml) sensitivitiesof 83%, 81% and 93% in detecting tumours< 5 mm, 5–30 mm and > 30 mm respectively.Unevaluable testsNone of the NMP22 studies specifically reportedthis information.CytologyCharacteristics of the includedstudiesA description of each of the 56 included cytologystudies is given in Appendix 12, which contains thecharacteristics of all of the included biomarker andcytology studies listed alphabetically by surname offirst author. Table 18 shows summary informationfor the 56 cytology studies.A total of 47 studies, reported in 56 papers, werediagnostic cross-sectional studies, 45,80,97,98,100–103,105,109–129,131–133,135,136,139–141, 145,146,148–153,165–174of which 11 45,101,109,110,112,116,118,126,127,165,174reported consecutiverecruitment. Nine studies 107,108,155,157–159,162–164 werecase–control studies and 11 studies 103,108,111,116,126,127,135,149,165,170,174were multicentre (Table 19).The 56 studies enrolled 22,260 participants,with 19,219 included in the analysis. Eightstudies 80,114,120,121,151,155,167,171 involving at least 872patients reported bladder wash cytology. In 46studies 45,80,97,101–103,105,107–114,116,118,122–124,126–129,131,132,135,136,139–141,148–151,153,158,159,162,164,165,168,170–172,1747888participants (45%) presented with a suspicion ofbladder cancer and 9487 (55%) had previouslydiagnosed bladder cancer. In 10 103,118,126,135,141,153,162,164,168,172of these studies the whole patientpopulation analysed (n = 4290) had a suspicionof bladder cancer and in 11 102,108,110,111,113,123,127,129,131,170,174the whole population analysed hadpreviously diagnosed bladder cancer (n = 5710).In total, 10 studies 98,100,119–121,155,157,163,166,167 did notreport this information.

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4TABLE 18 Summary of the characteristics of the cytology studiesCharacteristic Number Number of studiesStudy designCross-sectional diagnostic study 19,842 47Case–control 2418 9PatientsEnrolled 22,260 56Analysed 19,219Suspicion of or previously diagnosed BCSuspicion of BC 7888 (45%) 46 (82%)Previously diagnosed BC 9487 (55%)Not reported 3057 10 (18%)AgeMedian (range) of means (years) 67 (54 to 73) 33 (59%)Not reported – 23 (41%)SexMen 9702 (73%) 36 (64%)Women 3639 (27%) 20 (36%)Not reported 8578BC, bladder cancer.TABLE 19 Multicentre cytology studiesStudyBastacky 1999 165 3Grossman 2005 126 23Grossman 2006 127 23Karakiewicz 2006 170 10Messing 2005 111 4Miyanaga 1999 135 13Piaton 2003 116 19Raitanen 2002 174 18Sarosdy 2002 108 21Sarosdy 2006 103 23Stampfer 1998 149 3Number of centresAcross 33 studies 80,97,101–103,107,109,112–114,116,123,124,126–129,131,136,140,141,149–151,153,158,159,162,166,170–172,174providinginformation on patient age for the whole studypopulation, the median (range) of means was67 years (54 to 73 years). A total of 36 studiesprovided information on the gender of 13,341participants, of whom 9702 (73%) were men and3639 (27%) were women.© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.In total, 30 studies 80,102,103,108,111–114,118,119,121,122,126,127,135,136,139,150,151,153,155,158,159,162,164–168,174gave details ofwhen they took place, with an earliest start date of1990 165 and latest end date of July 2007. 118 Fifteenstudies took place in the USA, 97,103,105,108,111,126,127,129,139,148,149,153,158,159,165seven in Germany, 80,98,107,118,132,168,171four in the UK, 45,141,166,172 three eachin Italy 113,122,123 and Japan, 135,136,164 two eachin Austria, 112,151 Spain, 128,161 Sweden 114,155and India 131,150 and one each in Belgium, 167Finland, 174 France, 116 Greece, 124 Switzerland, 121 theNetherlands, 102 Turkey, 163 Canada 119 and SouthKorea, 157 while seven had multinational settings,with three taking place in Austria/Italy 101,109,110and the others taking place in Germany/USA, 140USA/Belgium, 100 Saudi Arabia/USA 120 and Austria/Germany/Italy/Spain/Sweden/Switzerland/Egypt/Japan/Canada/USA. 170Methodological quality of theincluded studiesFigure 23 summarises the quality assessment forthe 56 cytology studies. The results for the qualityassessment of the individual studies are shown inAppendix 13. In all studies the reference standard61

Results – biomarkers and cytology(cystoscopy with histological assessment of biopsiedtissue) was considered likely to correctly classifybladder cancer (Q2). In 55 studies (98%) thespectrum of patients who received the tests wasconsidered to be representative of those who wouldreceive the test in practice (Q1), incorporation biaswas avoided (Q6), uninterpretable test results werereported or there were none (Q10) and withdrawalsfrom the study were explained or there were none(Q11). In 54 studies (96%) partial verification biaswas avoided (Q4) and in 49 (88%) differentialverification bias was avoided (Q5). In 41 studies(73%) the time period between cytology and thereference standard was considered to be shortenough (1 month or less) to be reasonably surethat the patient’s condition had not changed inthe intervening period (Q3), in 40 studies (71%) aprespecified cut-off value for a positive test resultwas stated (Q12) and in 37 studies (66%) a cleardefinition of what was considered to be a positiveresult was provided (Q13).However, in 48 studies (86%) it was unclearwhether clinical review bias had been avoided(Q9), in 40 studies (71%) it was unclear whetherdiagnostic review bias had been avoided (Q8)and in 31 studies (55%) it was unclear whethertest review bias had been avoided (Q7). A total of53 studies (95%) did not report information onobserver variation in interpretation of test results(Q14).Assessment of diagnosticaccuracyPatient-level analysisA total of 36 studies 45,80,97,100,101,107,109,111,112,116,118,119,122–124,126–128,131,132,135,136,139–141,148,150,151,153,157,159,164,166,170,172,174reporting voided urine cytology, enrolling15,161 participants with 14,260 included in theanalysis, provided sufficient information to allowtheir inclusion in the pooled estimates for patientlevelanalysis. Figure 24 shows the sensitivityand specificity of the individual studies, pooledestimates and SROC curve for cytology patientbaseddetection of bladder cancer. The pooledsensitivity and specificity (95% CI) were 44% (38%to 51%) and 96% (94% to 98%), respectively, andthe DOR value (95% CI) was 19 (11 to 27). Acrossthe 36 studies the sensitivity for cytology rangedfrom 7% 136 to 100%, 172 and specificity ranged from78% 80 to 100%. 135 The median (range) PPV acrossstudies was 80% (27% to 100%) and the median(range) NPV was also 80% (38% to 100%).Most of the included studies in the pooledestimates contained a mixture of patients witha suspicion of bladder cancer and those withpreviously diagnosed bladder cancer. In sevenstudies 118,126,135,141,153,164,172 all of the participants(n = 3331) had a suspicion of bladder cancer[median (range) sensitivity and specificity acrossstudies 44% (16% to 100%) and 99% (87% to100%) respectively]. In six studies 111,123,127,131,170,174Spectrum representative?Reference standard correctly classifies condition?Time period between tests short enough?Partial verification bias avoided?Differential verification bias avoided?Incorporation bias avoided?Test review bias avoided?Diagnostic review bias avoided?Clinical review bias avoided?Uninterpretable results reported?Withdrawals explained?Prespecified cut-off?Positive result clearly defined?Data on observer variation reported?YesUnclearNo0 2040 60 80 100PercentageFIGURE 23 Summary of quality assessment of cytology studies (n = 56).62

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Sensitivity and specificity: individual study resultsStudy IDCasetta 2000 122Chalhal 2001a 166Chahal 2001b 45Del Nero 1999 123Giannopoulos 2000 124Grossman 2005 126Grossman 2006 127Gutierrez Banos 2001 128Halling 2000 97Karakiewicz 2006 170Kumar 2006 131Lahme 2001 132Lee 2001 157Lodde 2003 109May 2007 107Meiers 2007 100Messing 2005 111Mian 1999 112n19628521110514712876501501182542131109106225166624326249Sens %734924473816127058454145564171732347Spec %809497839299979398959693899484879398Study IDMian 2003 101Miyanaga 1999 135Miyanaga 2003 136Piaton 2003 116Ponsky 2001 139Potter 1999 172Poulakis 2001 140Raitanen 2002 174Ramakumar 1999 159Sadd 2002 141Schmitz-Drager 2008 118Sharma 1999 148Takeuchi 2004 164Talwar 2007 150Tetu 2005 119Tritschler 2007 80Wiener 1998 151Zippe 1999 153n18130913765160833673944111212028027866919687085291330Sens %455576262100623544484456442129445933Spec %9410098858599969095879693100999878100100SROC plotNumber of studiesSensitivity % (95% CI)Specificity % (95% CI)Positive likelihood ratioNegative likelihood ratioDOR (95% CI)Pooled analysis3644 (38 to 51)96 (94 to 98)10.8 (6.7 to 15.1)0.58 (0.51 to 0.64)18.6 (11.0 to 26.6)Sensitivity1.00.90.80.70.60.50.40.30.20.10.00.00.10.20.30.4 0.5 0.61–Specificity0.70.80.91.00.90.80.70.60.50.40.30.20.10.01.0SensitivityFIGURE 24 Cytology patient-level analysis: sensitivity, specificity, pooled estimates and SROC curve.all of the participants (n = 4195) had previouslydiagnosed bladder cancer [median (range)sensitivity and specificity across studies 38% (12%to 47%) and 94% (83% to 97%) respectively]. Fourstudies 100,119,157,166 did not report this information.Specimen-level analysisEight studies, 102,110,113,120,129,149,168,171 with at least1143 patients included in the analysis, reportedspecimen-level analysis (n = 3487) of voided urinecytology. (The study by Mian and colleagues 113enrolled 942 patients but did not report thenumber analysed, and in the study by Planz andcolleagues 171 it was unclear how many patientsunderwent voided urine cytology and howmany underwent bladder wash cytology.) Across© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.these studies the median (range) sensitivity andspecificity were 42% (38% to 76%) and 94% (58% to99%) respectively.Cytology using bladder washEight studies 80,114,120,121,151,155,167,171 involving at least872 patients reported bladder wash cytology.(It was unclear in the studies by Boman andcolleagues 155 and Planz and colleagues 171 how manypatients the specimen-based analysis related to.)Across four studies 80,114,121,151 reporting patientbaseddetection of bladder cancer (n = 608) themedian (range) sensitivity and specificity were 58%(53% to 76%) and 90% (62% to 100%) respectively(Olsson and colleagues 114 did not report specificity).This compares with 44% (95% CI 38% to 51%)63

Results – biomarkers and cytologysensitivity and 96% (95% CI 94% to 98%) specificityfor the 36 voided urine cytology studies included inthe pooled estimates.Across four studies 120,155,167,171 reporting specimenbaseddetection of bladder cancer (n = at least1076) the median (range) sensitivity and specificitywere 50% (38% to 62%) and 94% (83% to 99%)respectively. (Bhuiyan and colleagues 120 reportedsensitivity and specificity but not the number ofspecimens upon which this was based, and Olssonand colleagues 114 did not report specificity.) Thiscompares with a median (range) sensitivity of 42%(38% to 76%) and specificity of 94% (58% to 99%)across the eight studies reporting specimen-basedanalysis for voided urine cytology.All of the studies reporting bladder wash cytologycontained a mixture of patients with a suspicionof bladder cancer or previously diagnosed bladdercancer, or did not report numbers for these groupsof patients.Stage/grade analysisStudies reporting the sensitivity of cytology in thedetection of stage and grade of tumour categorisedthis information in different ways, including pTa,pTaG1, pTaG1–2, PTaG2, pTaG3, pTa pT1, pTapT1 CIS, pTa+CIS, ≥ pTa+CIS, pTa pT1G3,pTaG3–pT1, G1, G2, G1–2, pT1, pT1G1, pT1G2,pT1G1–2, pT1G3, pT1G3+CIS, pT1–T3b, pT1–4,CIS, CIS–pT1, G3, pT2, pT2 pT2a, pT2G2,pT2G3, pT2–3, pT2–4, ≥ pT2, pT3, pT3a 3b,pT3G3 and pT4 (see Appendix 14). If the numberof specimens included in the analysis was one perpatient then this was considered as a patient-basedanalysis.Table 20 shows the median (range) sensitivityof voided urine cytology, for both patient- andspecimen-based detection of tumours, within thebroad categories of less aggressive/lower risk andmore aggressive/higher risk (including CIS), andalso separately for CIS.Less aggressive, lower risk tumours(pTa, G1, G2)A total of 29 studies 45,97,100,101,103,107–109,111,112,116,119,123,124,126–128,131,132,140,141,150,151,157,159,164,166,170,174involving12,566 patients reported the sensitivity of voidedurine cytology for the patient-based detection ofless aggressive, lower risk tumours. Across thesestudies the median (range) sensitivity of cytologywas 27% (0% to 93%) (Table 20). Across threestudies 102,113,168 reporting the sensitivity of voidedurine cytology for specimen-based detection of lessaggressive, lower risk tumours (469+ participants,2411 specimens), the median (range) sensitivity was27% (8% to 78%).TABLE 20 Sensitivity of voided urine cytology in detecting stage/grade of tumourCytology sensitivity (%),median (range)Numberof patients(specimens) aNumber of studiesLess aggressive/lower riskPatient-based detection 27 (0 to 93) 12,566 29Specimen-based detection b 27 (8 to 78) 469+ (2411) 3More aggressive/higher risk including CISPatient-based detection 69 (0 to 100) 12,566 29Specimen-based detection b 79 (68 to 93) 608+ (3003) 4CISPatient-based detection 78 (0 to 100) 6870 17Specimen-based detection b 81 (76 to 93) 513+ (2895) 3a The numbers of patients and specimens are the totals included in the overall analysis by the studies.b Specimen-based detection: 469+, 608+, 513+. The ‘+’ represents the study by Mian and colleagues, 113 in which 942participants were enrolled but it was unclear how many were included in the analysis.64

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4More aggressive, higher risk tumours(pT1, G3, CIS)A total of 29 studies 45,97,100,101,103,107–109,111,112,116,119,123,124,126–128,131,132,140,141,150,151,157,159,164,166,170,174involving12,566 patients reported the sensitivity of voidedurine cytology for the patient-based detectionof more aggressive, higher risk tumours. Acrossthese studies the median (range) sensitivity ofcytology was 69% (0% to 100%) (Table 20). Acrossfour studies 102,113,167,168 reporting the sensitivity ofvoided urine cytology for specimen-based detectionof more aggressive, higher risk tumours (608+participants, 3003 specimens), the median (range)sensitivity was 79% (68% to 93%).Carcinoma in situA total of 17 studies 97,101,107–109,111,112,116,119,124,126,127,140,141,150,159,174involving 6870 patients reported thesensitivity of voided urine cytology for patientbaseddetection of CIS. Across these studies themedian (range) sensitivity of cytology was 78% (0%to 100%). Across three studies 113,167,168 reporting thesensitivity of voided urine cytology for specimenbaseddetection of CIS (513+ participants, 2895specimens), the median (range) sensitivity was 81%(76% to 93%).Number of tumoursThree studies reported the sensitivity of cytology indetecting bladder cancer in patients with varyingnumbers of tumours. Poulakis and colleagues 140 ina study involving 739 patients reported cytologysensitivities of 48%, 68% and 86% in patients withone, two to three, and more than three tumoursrespectively. Raitanen and colleagues 174 in a studyinvolving 570 patients reported on a subgroup of129 patients with no previous history of bladdercancer in which cytology sensitivities were 57%,54% and 71% in patients with one, two and morethan three tumours respectively. Takeuchi andcolleagues 164 in a study involving 669 patientsreported cytology sensitivities of 33%, 30% and82% in patients with one, two to four, and five ormore tumours respectively.Size of tumoursThree studies reported the sensitivity of cytology indetecting bladder cancer in patients with varyingsizes of tumours. Boman and colleagues 155 in astudy involving 250 patients reported cytologysensitivities of 35%, 33%, 55% and 87% indetecting new tumours ≤ 10 mm, 11–20 mm, 21–30 mm and > 30 mm, respectively, and 30%, 91%and 100% in detecting recurrent tumours ≤ 10 mm,11–20 mm and > 21 mm respectively. Messingand colleagues 111 in a study involving 326 patients© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.reported cytology sensitivities of 18%, 26% and20% in detecting tumours < 10 mm, 10–30 mm and> 30 mm respectively. Takeuchi and colleagues 164in a study involving 669 patients reported cytologysensitivities of 21%, 47% and 75% in detectingtumours < 10 mm, 10–30 mm and > 30 mmrespectively.Unevaluable testsSix studies 101,103,114,118,119,174 specifically reportedunevaluable tests. Overall, 54 of 2566 tests (2%)could not be evaluated. Across studies, the median(range) percentage of tests that were unevaluablewas 1% (0.6% to 4%).Observer variationTwo studies reported observer variation.Hughes and colleagues 129 reported that all 128specimens were independently reviewed by twocytopathologists, who were approximately 80%concordant in their interpretation of the cases.In the case of approximately 20% of specimensabout which there was disagreement concerningthe cytological diagnosis, the cytospin wasreviewed by the two pathologists simultaneouslyand an agreement was reached. 129 Sarosdy andcolleagues 108 reported that local site results wereavailable in 43 cases and there was agreementwith study central cytology in 36 (84%). Of theremaining seven cases, four were positive at the siteand negative at the study testing laboratory, andthree were negative at the investigation site andpositive at the study testing laboratory. 108 Study sitecytology was available in three cases of CIS andeight cases of G3 tumour, with 100% agreementbetween study site and central laboratorycytopathology interpretation in these 11 cases. 108Studies directly comparingtestsFISH versus cytologyFive 97,98,100,101,107 of the studies included in thepooled estimates for FISH and for cytology directlycompared the two tests. The studies enrolled1377 participants, with 1119 included in theanalysis for FISH and 1198 for cytology. Figure25 shows the sensitivity and specificity of theindividual studies, pooled estimates and SROCcurves for these five studies. The pooled estimate(95% CI) for the sensitivity of FISH was 81%(66% to 97%) compared with 54% (39 to 80%) forcytology, whereas the pooled estimate (95% CI)for the specificity of FISH was 82% (68% to 97%)compared with 92% (84% to 99%) for cytology.65

Results – biomarkers and cytologyImmunoCyt versus cytologySix 101,109,111,112,116,118 of the studies included in thepooled estimates for ImmunoCyt and for cytologydirectly compared the two tests. The studiesenrolled 2016 participants, with 1912 includedin the analysis. Figure 26 shows the sensitivityand specificity for the individual studies, pooledestimates and SROC curves for these six studies.The pooled estimate (95% CI) for the sensitivity ofImmunoCyt was 82% (76% to 89%) compared with44% (35% to 54%) for cytology, whereas the pooledestimate (95% CI) for the specificity of ImmunoCytwas 85% (71% to 85%) compared with 94% (91% to97%) for cytology.NMP22 versus cytologyIn total, 16 45,80,123,126–128,131,132,139–141,148,150,151,153,159 ofthe studies included in the pooled estimates forNMP22 and for cytology directly compared thetwo tests. The studies enrolled 5623 participants,with 5563 included in the analysis for NMP22 and5402 for cytology. Figure 27 shows the sensitivityand specificity for the individual studies, pooledestimates and SROC curves for these 16 studies.The pooled estimate (95% CI) for the sensitivityof NMP22 was 70% (59% to 80%) compared with40% (31% to 49%) for cytology, whereas the pooledestimate (95% CI) for the specificity of NMP22 was81% (74% to 88%) compared with 97% (95% to99%) for cytology.Studies reportingcombinations of testsIn total, 16 studies reported the sensitivityand specificity of combinations of tests indetecting bladder cancer, including FISH andcytology, 94,102 FISH and cystoscopy, 99 ImmunoCytand cytology, 101,109–113,116,119 ImmunoCyt andcystoscopy, 118 NMP22 and cytology, 131,164NMP22 and cystoscopy 126,127 and cytology andcystoscopy. 118,127 Although not explicitly stated inthe reports, the definition of a positive test resultfor the combined tests was a positive result oneither of the tests included in the combination.The exception to this was the study by Daniely andcolleagues, 94 which reported the test performanceof FISH combined with cytology.FISH and cytologyTwo studies 94,102 reported the sensitivity andspecificity of FISH and cytology used incombination. In a patient-level analysis (n = 115),Daniely and colleagues 94 reported sensitivity andspecificity of 100% and 50%, respectively, for FISHand cytology used in combination (results werenot presented separately for the individual tests).A test was reported as positive if at least one cellabnormality in both cytology and FISH was found.In the case of abnormal FISH and normal cytology,a minimum of four cells with a gain of two or moreSensitivity and specificity of studies comparing FISH and cytologySROC plot of FISH and cytologyStudy IDHalling 2000 97Junker 2006 98May 2007 107Meiers 2007 100Mian 2003 101Number of studiesSensitivity % (95% CI)Specificity % (95% CI)Positive likelihood ratioNegative likelihood ratioDOR (95% CI)n15112116662457FISHSens %8160539396Spec %9681749045n118109166624181Pooled analysis of FISH and cytologyFISH581 (66 to 97)82 (68 to 974.6 (0.9 to 8.3)0.23 (0.04 to 0.42)20.1 (

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Sensitivity and specificity of studies comparing ImmunoCyt and cytologySROC plot of ImmunoCyt and cytologyStudy IDnLodde 2003 109 225Messing 2005 111 326Mian 1999 112 249Mian 2003 101 181Piaton 2003 116 651Schmitz-Drager 2802008 118ImmunoCytSens %878186867385Spec %677579718288n225326249181651280CytologySens %412347456244Spec %949398948596Pooled analysis of ImmunoCyt and cytologyImmunoCyt CytologyNumber of studiesSensitivity % (95% CI)Specificity % (95% CI)Positive likelihood ratioNegative likelihood ratioDOR (95% CI)682 (76 to 89)85 (71 to 85)3.8 (2.7 to 4.9)0.22 (0.15 to 0.30)17.0 (9.5 to 24.5)644 (35 to 54)94 (91 to 97)7.2 (4.0 to 10.5)0.59 (0.50 to 0.69)12.2 (6.1 to 18.3)Sensitivity1.00.90.80.70.60.50.40.30.20.10.00.0 0.1 0.2 0.3Test: / Cytology/ Immunocyt0.4 0.5 0.61–Specificity0.70.80.91.00.90.80.70.60.50.40.30.20.10.01.0SensitivityBIC difference between this and simpler model: 183.2–197.2 = –14FIGURE 26 ImmunoCyt vs cytology – patient-level analysis: sensitivity, specificity, pooled estimates and SROC curve.chromosomes or 12 or more cells with homozygousloss of the 9p21 locus was required for a positivediagnosis. The study by Moonen and colleagues 102involving 105 patients reported a specimen-basedanalysis (n = 103), with sensitivity and specificity of39% and 90%, respectively, for FISH, 41% and 90%for cytology and 53% and 79% for the tests used incombination.FISH and cystoscopyIn a patient-based analysis, Kipp and colleagues 99in a study involving 124 patients reported thesensitivity and specificity of FISH and cystoscopy(not stated whether flexible or rigid) used incombination. They reported sensitivity andspecificity of 62% and 87%, respectively, for FISH,67% and 85% for cystoscopy and 87% and 79% forthe tests used in combination. A definition of whatconstituted a positive test result for the combinedtests was not given.ImmunoCyt and cytologyEight studies reported sensitivity and specificityfor the tests of ImmunoCyt and cytology used incombination. Six studies 101,109,111,112,116,119 involving1997 patients reported patient-based detection andtwo studies 110,113 involving 1137 patients reportedspecimen-based detection (2220 specimens).© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.The median (range) sensitivities and specificitiesof ImmunoCyt, cytology, and ImmunoCyt andcytology across the studies reporting patient- andspecimen-based detection are shown in Table 21.The sensitivity of the tests in combination for bothpatient- and specimen-based detection (87% and88% respectively) was slightly higher than that ofImmunoCyt alone (84% and 78%), whereas thespecificity (68% and 76%) was much lower than thatof cytology alone (94% and 97%).ImmunoCyt and cystoscopyIn a patient-based analysis (n = 280), Schmitz-Drager and colleagues 118 reported sensitivityand specificity of 85% and 88%, respectively, forImmunoCyt, 84% and 98% for cystoscopy (notstated whether flexible or rigid) and 100% and 87%for the tests used in combination.NMP22 and cytologyIn a patient-based analysis, two studies 131,164involving 800 patients reported the sensitivityand specificity of NMP22 and cytology used incombination. The study by Kumar and colleagues 131involving 131 patients used the NMP22BladderChek point of care test with a cut-off of10 U/ml. They reported sensitivity and specificityof 85% and 78%, respectively, for NMP22, 41%67

Results – biomarkers and cytologyStudy IDChahal 2001b 45Del Nero 1999 123Grossman 2005 126Grossman 2006 127Gutierrez Banos 2001 128Kumar 2006 131Lahme 2001 132Ponsky 2001 139Poulakis 2001 140Ramakumar 1999 159Saad 2002 141Sharma 1999 148Talwar 2007 150Tritschler 2007 80Wiener 1998 151Zippe 1999 153Sensitivity and specificity of studies comparing NMP22 and cytologyNMP22Cytologyn2111051331668150131109608739196120278196100291330Sens %338356507685638879538156676548100Spec %92878687917861847060879381406986n211105128765015013110960873911212027819685291330Sens %24471612704145626244482921445933Spec %97839997939693859695871009978100100Number of studiesSensitivity % (95% CI)Specificity % (95% CI)Positive likelihood ratioNegative likelihood ratioDOR (95% CI)Pooled analysis of NMP22 and cytologyNMP221870 (59 to 80)81 (74 to 88)3.6 (2.3 to 5.0)0.38 (0.25 to 0.50)9.6 (3.8 to 15.4)Cytology1840 (31 to 49)97 (95 to 99)12.2 (4.3 to 20.2)0.62 (0.53 to 0.71)19.8 (5.9 to 33.7)BIC difference between this and simpler model: 511.5–526.7 = –15.2Sensitivity1.00.90.80.70.60.50.40.30.20.10.00.0Test: //0.1SROC plot of NMP22 and cytology0.2CytologyNMP220.30.4 0.5 0.61–Specificity0.70.80.91.00.90.80.70.60.50.40.30.20.10.01.0SensitivityFIGURE 27 NMP22 vs cytology – patient-level analysis: sensitivity, specificity, pooled estimates and SROC curve.TABLE 21 Median (range) sensitivity and specificity across studies reporting ImmunoCyt plus cytologyTest Sensitivity (%), median (range) Specificity (%), median (range)Patient-based detection (n = 6 studies)ImmunoCyt 84 (73 to 87) 73 (62 to 82)Cytology 43 (23 to 62) 94 (85 to 98)ImmunoCyt + cytology 87 (81 to 90) 68 (61 to 79) aSpecimen-based detection (n = 2 studies)ImmunoCyt 78 (71 to 85) 76 (73 to 78)Cytology 44 (39 to 49) 97 (95 to 99)ImmunoCyt + cytology 88 (86 to 89) 76 (73 to 78)68a The median (range) specificity for ImmunoCyt + cytology is based on five studies as Piaton and colleagues 116 did notreport specificity for the tests in combination.

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4and 96% for cytology and 91% sensitivity forthe tests used in combination (specificity wasnot reported). 131 The study by Takeuchi andcolleagues 164 involving 669 patients used a cut-offof 12 U/ml for NMP22. They reported sensitivityand specificity of 58% and 80%, respectively,for NMP22, 44% and 100% for cytology and60% sensitivity for the tests used in combination(specificity was not reported). In both studies thesensitivity for the tests in combination was slightlyhigher than that for NMP22 alone, although therewas a wide difference in the sensitivity values forNMP22 reported by the two studies.NMP22 and cystoscopyIn a patient-based analysis (n = 1999), two studiesby Grossman and colleagues 126,127 reportedthe sensitivity and specificity of the NMP22BladderChek point of care test and cystoscopy(not stated whether flexible or rigid) used incombination. Both studies used a cut-off of 10 U/ml to define a positive NMP22 test result. In thefirst study 126 sensitivity was 56% and specificity 86%for NMP22 (1331 patients), whereas in 79 patientsdiagnosed with bladder cancer the sensitivity ofcystoscopy and the tests used in combination was89% and 94% respectively. In the second study 127sensitivity was 50% and specificity 87% for NMP22(668 patients), whereas in 103 patients diagnosedwith bladder cancer the sensitivity of cystoscopyand the tests used in combination was 91% and99% respectively.Cytology and cystoscopyIn a patient-based analysis (n = 280), Schmitz-Drager and colleagues 118 reported sensitivityand specificity of 44% and 96%, respectively, forcytology, 84% and 98% for cystoscopy (not statedwhether flexible or rigid) and 88% and 95% forthe tests used in combination. In 103 patientsdiagnosed with bladder cancer Grossman andcolleagues 127 reported sensitivity of 12% forcytology, 91% for cystoscopy and 94% for the testsused in combination.SummaryA total of 71 studies, published in 83 reports, metthe inclusion criteria for studies reporting the testperformance of biomarkers (FISH, ImmunoCyt,NMP22) and cytology in detecting bladder cancer.In total, 14 studies enrolling 3321 participantsreported on FISH, 10 studies enrolling 4199© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.participants reported on ImmunoCyt, 41 studiesenrolling 13,885 participants reported on NMP22and 56 studies enrolling 22,260 participantsreported on cytology. The vast majority of thestudies were diagnostic cross-sectional studies(n = 59, 83%), with the remainder being case–control studies (n = 12, 17%).Pooled estimates with 95% CIs for sensitivity,specificity, positive and negative likelihood ratiosand DORs for each of the tests were undertakenfor patient-level analysis. Table 22 shows thepooled estimates for sensitivity, specificity andDOR for each of the tests. Sensitivity was highestfor ImmunoCyt at 84% (95% CI 77% to 91%) andlowest for cytology at 44% (95% CI 38% to 51%).ImmunoCyt (84%, 95% CI 77% to 91%) had highersensitivity than NMP22 (68%, 95% CI 62% to 74%),with the lack of overlap of the CIs supportingevidence of a difference in sensitivity between thetests in favour of ImmunoCyt. FISH (76%, 95% CI65% to 84%), ImmunoCyt (84%, 95% CI 77% to91%) and NMP22 (68%, 95% CI 62% to 74%) allhad higher sensitivity than cytology (44%, 95% CI38% to 51%), and again the lack of overlap betweenthe biomarker and cytology CIs supportingevidence of a difference in sensitivity in favour ofthe biomarkers over cytology.Although sensitivity was highest for ImmunoCytand lowest for cytology, this situation was reversedfor specificity, which was highest for cytologyat 96% (95% CI 94% to 98%) and lowest forImmunoCyt at 75% (95% CI 68% to 83%). Cytology(96%, 95% CI 94% to 98%) had higher specificitythan FISH (85%, 95% CI 78% to 92%), ImmunoCyt(75%, 95% CI 68% to 83%) or NMP22 (79%, 95%CI 74% to 84%), with the lack of overlap betweenthe cytology and biomarker CIs supportingevidence of a difference in specificity in favour ofcytology over the biomarkers.DORs (95% CI) ranged from 8 (5 to 11) to 19 (6 to26), with higher DORs indicating a better ability ofthe test to differentiate between those with bladdercancer and those without. Based on the DORvalues, FISH and cytology performed similarlywell [18 (3 to 32) and 19 (11 to 27) respectively],ImmunoCyt slightly less so [16 (6 to 26)] andNMP22 relatively poorly [8 (5 to 11)]. However, asthe DOR CIs for each of the tests all overlappedthese results should be interpreted with caution.Across studies the median (range) PPV was highestfor cytology at 80% (27% to 100%) and FISH at78% (27% to 99%), followed by ImmunoCyt at69

Results – biomarkers and cytology70TABLE 22 Summary of pooled estimate results for biomarkers and cytology for patient-based detection of bladder cancerTest Number of studies Number analysed Common cut-off Sensitivity (%) (95% CI) Specificity (%) 95% CI) DOR (95% CI)FISH 12 2535 Gain of more than one or more 76 (65 to 84) 85 (78 to 92) 18 (3 to 32)than two chromosomes a84 (77 to 91) 75 (68 to 83) 16 (6 to 26)ImmunoCyt 8 2896 At least one green or one redfluorescent cellNMP22 28 10,119 ≥ 10 U/ml 68 (62 to 74) 79 (74 to 84) 8 (5 to 11)Cytology 36 14,260 Cytologist subjective judgement 44 (38 to 51) 96 (94 to 98) 19 (11 to 27)a FISH, common cut-off – see Appendix 16 for a detailed description of the cut-offs used by each of the FISH studies.TABLE 23 Summary of median (range) sensitivity of tests across studies for patient-level detection of stage/grade of bladder cancerNumberof studies CIS, median (range)(patients) a sensitivity across studiesNumber More aggressive/higher risk(patients) a sensitivity across studiesof studies including CIS, median (range)Number Less aggressive/lower risk,(patients) a across studiesof studies median (range) sensitivityTestFISH 10 (2164) 65 (32 to 100) 10 (2164) 95 (50 to 100) 8 (1067) 100 (50 to 100)ImmunoCyt 6 (2502) 81 (55 to 90) 6 (2502) 90 (67 to 100) 6 (2502) 100 (67 to 100)NMP22 18 (4685) 50 (0 to 86) 19 (7556) 83 (0 to 100) 11 (3453) 83 (0 to 100)Cytology 29 (12,566) 27 (0 to 93) 29 (12,566) 69 (0 to 100) 17 (6870) 78 (0 to 100)a The number of patients refers to the number included in the overall analysis by the studies.TABLE 24 Pooled estimates for sensitivity and specificity for tests being directly compared within studiesSpecificity (%)(95% CI)Sensitivity (%)(95% CI)Specificity (%)(95% CI) TestSensitivity (%)(95% CI)Number of studies(patients) a TestComparisonFISH vs cytology 5 (1119; 1198) FISH 81 (66 to 79) 82 (68 to 97) Cytology 54 (39 to 80) 92 (84 to 99)ImmunoCyt vs cytology 6 (1912; 1912) ImmunoCyt 82 (76 to 89) 85 (71 to 85) Cytology 44 (35 to 54) 94 (91 to 97)NMP22 vs cytology 16 (5563; 5402) NMP22 70 (59 to 80) 81 (74 to 88) Cytology 40 (31 to 49) 97 (95 to 99)a The numbers in parentheses separated by a semicolon represent the number of patients included in the analysis for each of the tests being compared, e.g. (1119; 1198), 1119patients were included in the analysis for FISH, 1198 were included in the analysis for cytology.

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 454% (26% to 70%) and NMP22 at 52% (13% to94%). The median (range) NPV was highest forImmunoCyt at 93% (86% to 100%), followed byFISH at 88% (36% to 97%), NMP22 at 82% (44% to100%) and cytology at 80% (38% to 100%).Table 23 summarises the sensitivity of the testsin detecting stage/grade of tumour. ImmunoCythad the highest median sensitivity across studies(81%) for detection of less aggressive/lower risktumours whereas FISH had the highest mediansensitivity across studies (95%) for detection ofmore aggressive/higher risk tumours. For detectionof CIS the median sensitivity across studies for bothFISH and ImmunoCyt was 100%. Cytology hadthe lowest sensitivity across studies for detectingless aggressive/lower risk tumours (27%), moreaggressive/higher risk tumours (69%) and also CIS(78%). The median sensitivity across studies foreach test was consistently higher for the detectionof more aggressive/higher risk tumours than itwas for the detection of less aggressive, lower risktumours.Some of the studies included in the pooledestimates for the individual tests also directlycompared tests, e.g. FISH versus cytology. Table24 shows the pooled estimates for sensitivity andspecificity for those tests being directly comparedin studies and reporting a patient-level analysis.In each set of comparisons cytology had lowersensitivity but higher specificity than the biomarkerwith which it was being compared. ImmunoCythad a statistically significant higher sensitivity(82%, 95% CI 76% to 89%) than that of cytology(44%, 95% CI 35% to 54%), whereas cytology had astatistically significant higher specificity (94%, 95%CI 91% to 97%) than that of ImmunoCyt (85%,95% CI 71% to 85%). Similarly, NMP22 had astatistically significant higher sensitivity (70%, 95%CI 59% to 80%) than that of cytology (40%, 95% CI31% to 49%), whereas cytology had a statisticallysignificant higher specificity (97%, 95% CI 95% to99%) than that of NMP22 (81%, 95% CI 74% to88%).In studies reporting the sensitivity and specificity oftests used in combination, sensitivity was generallyhigher but specificity lower for the combinedtests compared with the higher value of theindividual tests. Most combinations of tests werereported by only one or two studies, apart from thecombination of ImmunoCyt and cytology, whichwas reported by eight studies.In studies specifically reporting unevaluable tests,rates were 6.1% (65/1059, five studies) for FISH,5% (279/5292, 10 studies) for ImmunoCyt and 2%(54/2566, six studies) for cytology. None of theNMP22 studies specifically reported unevaluabletests.71© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Chapter 6Assessment of cost-effectivenessUsing the care pathways described in Chapter 2,an economic model was developed to estimatethe cost-effectiveness of several managementstrategies for the initial diagnosis and follow-upof bladder cancer. This chapter describes how thedata to estimate cost-effectiveness were derived andhow these data were used in the economic model.The perspective adopted for the cost-effectivenessanalysis was that of the NHS.Economic model for initialdiagnosis and follow-up ofbladder cancerModel structureBased on the care pathway described in Chapter2, the model structure was developed followingconsultation with clinicians and taking intoconsideration the approaches adopted by theexisting economic evaluations 153,158,175–180 identifiedfrom the literature. The approach attemptsto model patients passing through the wholesequence of care and determine the overall impacton costs and the clinical consequences. Figure 28shows a simplified model structure for the primarydiagnosis and follow-up management of bladdercancer. Within this model people with suspectedbladder cancer will receive tests and investigationsto diagnose bladder cancer. Subsequentmanagement will depend upon the findings ofthese tests and the nature of any bladder cancerdetected. The absorbing state in the model is deathfrom either bladder cancer or other causes.The cost-effectiveness analysis was performed intwo parts. The first part considered the diagnostictests and consisted of a decision tree modelelement and the second part considered the followupof patients after diagnosis using a Markovmodel.Decision tree modelThe decision tree, constructed using TreeAgeSoftware, displays the temporal and logicalsequence of a clinical decision problem. Althoughthis decision tree does not explicitly specify thetime over which diagnosis takes as part of the© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.model structure, going from initial presentation tofinal diagnosis may take weeks or even months.As described in Chapters 1 and 2 there does notappear to be a single standard strategy in theUK. Flexible cystoscopy alone or combined withcytology followed by white light rigid cystoscopy arethe main diagnostic tests performed. Cytology orbiomarkers followed by WLC or PDD for the initialdiagnosis of bladder cancer are less commonly usedin the UK, but the use of cytology or biomarkersfollowed by WLC or PDD may be feasible. Theaim of this model is to reflect the costs andconsequences of these tests compared with one‘standard’ strategy, ‘flexible cystoscopy followed byWLC’.Interventions of diagnosis and follow-upThe interventions included in the model wereflexible cystoscopy, cytology, three types ofbiomarkers (NMP22, FISH, ImmunoCyt), WLCand PDD. Although flexible cystoscopy combinedwith cytology and a biomarker as the first suite oftests may be an option for the primary diagnosisof bladder cancer, there is little information aboutthe results of these tests used in combination, asreported in Chapters 4 and 5. Table 25 summarisesthe potential strategies that are considered in themodel. These options were based on advice fromclinical experts about strategies that are currentlyin use or those that can potentially be used.Strategies 1–6 consider the use of a single test forinitial diagnosis. These options might representsituations that clinical practice might movetowards although they may not be currently usedin practice. Strategies 7–16 represent alternativesituations in which two or more tests are used inthe initial phase of diagnosis. Across all strategiesthe choice of second level diagnostic test variesbetween WLC and PDD. The strategies also differin terms of the tests used for follow-up surveillance.In our study we have assumed that a single testis used for initial surveillance with any positivesconfirmed by WLC.It should be noted that none of our strategiesexplicitly considers the use of ultrasound.Ultrasound might be considered part of all of the73

Assessment of cost-effectivenessSymptoms ofbladder cancerNon-muscleinvasiveDiagnosisMuscleinvasiveManagementMetastasesDeathFIGURE 28 Model structure.strategies when the patient population is restrictedto haematuria. In such a situation this would haveno impact on incremental costs (as all patientsunder all strategies incur the test) although it mayalter the likelihood of subsequent testing.Figure 29 illustrates a simplified model structure forthe decision tree model for diagnosis of bladdercancer when a single test is used as part of theinitial diagnosis (i.e. strategies 1–6). Figure 30illustrates the model structure for the situation inwhich two tests are used as part of the initial testingTABLE 25 Diagnostic strategiesPrimary diagnosisFollow-up surveillanceInitial test Second test Initial test Second testStrategy CSC CTL BM WLC PDD CSC CTL BM WLC PDD1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 BM, biomarker; CSC, flexible cystoscopy; CTL, cytology.74

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4(strategies 7–14). When three tests are used incombination (strategies 15 and 16) a similar modelstructure to that in Figure 30 is developed (figurenot shown).In Figure 29 a patient may, for example, arrive in ahospital with symptoms of haematuria. Taking thepatient’s history and symptoms into account, thephysician may perform an invasive test (flexiblecystoscopy) or a non-invasive test (e.g. cytologyand biomarkers). The results of these tests couldbe either negative or positive. The negative testresult could be either a false or a true negative.If the first single test in Figure 29 is negative, it isassumed that there appears to be no evidence ofbladder cancer and the patient is deemed not tohave bladder cancer. If the result of the first testis positive (which might be either a true or a falsepositive) the patient will be further investigatedusing the second test, which will be either PDD orWLC. As with the first test there are four potentialtest results: true negative, false negative, truepositive and false positive. As there is a risk ofdeath associated with the use of general anaesthesiarequired for rigid cystoscopy, there is a chance thatthe patient may die whilst undergoing or as a resultof undergoing the second test.For the strategies in which two tests form part ofthe initial diagnosis (strategies 7–14) the first testthat a patient receives will be flexible cystoscopy(Figure 30). If the result is negative (it might beeither a true or a false negative) it is assumed thatthe patient will be further tested using cytology ora biomarker. If the result of cytology or a biomarkeris negative the patient will be deemed not to havebladder cancer. If the result of the first test ispositive (which might be either a true or a falsepositive) the patient will be further investigatedusing the second test, which will be either PDD orWLC. Patients who test positive with cytology ora biomarker will be handled in a similar manner.As with the first test there are four potential testresults: true negative, false negative, true positiveand false positive. As there is a risk of deathassociated with the use of general anaesthesiarequired for rigid cystoscopy, there is a chance thatthe patient may die whilst undergoing or as a resultof undergoing the second test.Strictly speaking, Figure 30 describes the situationin which only those negative on flexible cystoscopy(CSC) receive either cytology (CTL) or a biomarker(BM) test. In practice, because of the way thatservices might be organised, the different testsmay be performed during the same visit, i.e. thosewho are positive with flexible cystoscopy may alsoreceive either cytology or a biomarker test. Theimplications of this are that, given the cost dataavailable for this study, the average cost per patientin actual practice would be increased comparedwith the practice described in Figure 29 (there willbe no impact on effectiveness as all positives gothrough to the next level of testing). It should benoted that the practice of conducting additionaltests at the same time as flexible cystoscopy is likelyto be adopted because it is logistically easier toorganise, i.e. the real opportunity costs of currentpractice are less than would be predicted from theunit costs available for this study. For this reasonwe have assumed that a more realistic estimate ofcosts will be provided by a model following thestructure set out in Figure 30 but we have providedan additional analysis to illustrate the effect onPresentationwithmicroscopicor grosshaematuriaor lowerurinary tractsymptomsCSCCTLBMNegativePositiveWLCPDDNegativePositiveTruenegativeFalsenegativeTruenegativeFalsenegativeTruepositiveFalsepositiveFIGURE 29 Decision tree model structure for single diagnostic technology as the first test. BM, biomarker; CSC, flexible cystoscopy; CTL,cytology.75© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Assessment of cost-effectivenessMicroscopicor grosshaematuriaLowerurinary tractsymptomsCSCNegativePositiveCTLBMNegativePositiveWLCPDDNegativePositiveTruenegativeFalsenegativeTruenegativeFalsenegativeTruepositiveFalsepositiveFIGURE 30 Decision tree model structure for flexible cystoscopy combined with cytology or biomarker as the first test. BM, biomarker;CSC, flexible cystoscopy; CTL, cytology.76costs when two or more tests are conducted on allpatients presenting for initial diagnosis.Estimation of probabilities required forthe decision tree modelThe probabilities used to populate the decisionmodel were calculated according to the standardconventions of Bayes’ theorem. The essenceof the calculations is that, once the sensitivityand specificity of a test are known, along withthe a priori probability of disease, the posteriorprobabilities of disease and absence of diseasecan be determined. Accordingly, if a patienthas an abnormal test result, the probability ofdisease – the ‘true positive rate’, also referredto as the ‘positive predictive value’ (PPV) – isrepresented as p(BC+|T+), and if the patient hasa normal test result, the probability of disease –the ‘false-negative rate’ – is similarly presented asp(BC+|T–). These are calculated as follows:p(BC+|T+) = p(T+|BC+) p(BC+)/(p(T+|BC+) p(BC+) + p(T+|BC–) p(BC–))p(BC+|T–) = p(T–|BC+) p(BC+)/(p(T–|BC+)p(BC+) + p(T–|BC–)p(BC–))where BC = bladder cancer, T+ = test positive,T– = test negative, p(T+|BC+) = sensitivity,p(BC+) = prior probability of disease (prevalenceor incidence), p(T+|BC–) = 1 – specificity,p(BC–) = 1 – prevalence (or incidence), p(T–|BC+) = 1 – sensitivity and p(T–|BC–) = specificity.When two tests are connected in series, thecalculations are the same except that the priorprobability of disease (prevalence or incidence) forthe second test is the calculated ‘true positive rate’of the first test.To illustrate this in the construction and analysisof the bladder cancer primary diagnosis tree(Appendix 17, Figure 37), the strategy ‘flexiblecystoscopy (CSC) followed by WLC’ is considered.The probability of a test positive result followingflexible cystoscopy is:pPos_CSC = (Sens_CSC*priori) + (1 – Spec_CSC)*(1 – priori)where Sens_CSC = sensitivity of flexible cystoscopy,Spec_CSC = specificity of flexible cystoscopy andpriori is the prevalence or incidence rate forpatients with suspected bladder cancer before theflexible cystoscopy test.From this, the probability of a:• negative result for flexible cystoscopy is1 – pPos_CSC• false negative for flexible cystoscopy is pFN_CSC = (1 – Sens_CSC)*priori/((1 – Sens_CSC)*priori + Spec_CSC*(1 – priori))• true negative is 1 – pFN_CSC• positive result for WLC following apositive flexible cystoscopy result ispPos_CSC_WLC = (Sens_WLC*pPPV_CSC) + (1 – Spec_WLC)*(1 – pPPV_CSC),where Sens_WLC = sensitivity of WLC,Spec_WLC = specificity of WLC and pPPV_CSC = positive predictive value of flexiblecystoscopy = (Sens_CSC*priori)/pPos_CSC

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4• true positive for WLC following a positiveflexible cystoscopy result is pTP_CSC_WLC = (Sens_WLC*pPPV_CSC)/pPos_CSC_WLC• false positive for WLC following flexiblecystoscopy is 1 – pTP_CSC_WLC• false negative for WLC following flexiblecystoscopy is pFN_CSC_WLC = [Spec_WLC*(1 – pPPV_CSC)]/(1 – pPos_CSC_WLC)• true negative is 1 – pFN_CSC_WLC• the NPV after a negative result for CSC ispNPV_CSC = [Spec_CSC*(1 – priori)]/(1 – pPos_CSC).The probabilities for the remaining strategies inthe tree are calculated in a similar manner.It is important to quantify the false-positive andfalse-negative values for each strategy, as theseprovide valuable information to the clinician inaddition to the cost and number of true casesdetected. The implications of false-positive resultswithin the model are the cost of testing andtreating patients and the associated morbidity anddiscomfort of further investigation and treatment.False-positive results may also induce adversepsychological responses in patients in terms ofthe needless distress that a positive result mightcause and by leading to questioning of futureresults that are negative. In the case of falsenegativeresults the patient may have a serious orlife-threatening condition that is missed, resultingin a potentially poorer prognosis following latedetection, such as CIS missed by WLC, as well aspsychological distress from false reassurance. Inthe decision model patients with a false-negativeevaluation following the first (flexible cystoscopy,cytology or biomarkers) or second (PDD/WLC) testmay be subsequently correctly diagnosed as theircontinuing symptoms worsen. In the case of truenegative results, it is assumed that the patients willnot need further investigation.Management of bladder cancerPatients with true-positive results (confirmedbladder cancer) are classified into two types: nonmuscle-invasiveand muscle-invasive disease (Figure31). Those with muscle-invasive tumours will notbe discharged but are managed usually with eithersurgery (radical cystectomy) or radical radiotherapywith or without chemotherapy and routinechecking thereafter and treatment. All patientswith non-muscle-invasive tumours will undergoa follow-up test at 3 months after the primarydiagnosis because of the high chance of recurrenceand a chance of progression. For each risk groupthere are similar outcomes considered in initialdiagnosis: true positive, false positive, true negativeand false negative (Appendix 17, Figure 37).It is assumed that the first test used in the followupof patients will be the same as the test usedfor primary diagnosis and the second test will beWLC. To illustrate the construction and analysisof each risk group, strategy ‘flexible cystoscopy(CSC) followed by WLC in primary diagnosis andfollow-up by CSC’ is considered. In the case of eachgroup, the probability of:• true positive is pTP_Riskgroup = Sens_CSC*Recurrence rate of risk group at 3 months• true negative is pTN_Riskgroup = Spec_CSC*(1 – Recurrence rate of risk group at 3months)Low-risk groupNon-muscle invasiveIntermediate-risk groupFollow-up nonmuscleinvasiveBladder cancerMuscle invasiveHigh-risk groupLocally muscle invasiveMetastasesRadiotherapy orsurgery formuscle invasiveFIGURE 31 Classification of bladder cancer.77© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Assessment of cost-effectiveness• false negative is pTP_Riskgroup = (1 – Sens_CSC)* Recurrence rate of risk group at 3months• false positive is pFN_Riskgroup = (1 – Spec_CSC)*(1 – Recurrence rate of risk group at 3months).As described in the care pathway reported inChapter 2, bladder cancer treatment options willdepend on classification of disease (Table 26).To determine the efficiency of each strategy theterminal nodes (Appendix 17, Figure 37) of thetree were assigned a value of either ‘1’ or ‘0’. Thisenabled the following solutions to be calculated:mean cost per case detected – achieved byassigning the value ‘0’ to dead terminal node andthe value ‘1’ to the others.Markov modelAt the end of each branch of the decision tree thepatients will enter one of the predefined states ofthe Markov model (Appendix 17, Figures 36 and38). The health states within the Markov model areconsidered to reflect possible paths of recurrenceand progression of bladder cancer based oninformation of the primary diagnosis and followingthe follow-up visit carried out 3 months after initialtreatment of the bladder cancer.As indicated in the care pathways described inChapter 2, there are two elements in the Markovmodels: non-muscle invasive (TaT1) and muscleinvasive (T2 or > T2). In the case of muscleinvasivedisease, patients have a serious and lifethreateningcondition and high mortality andmorbidity rates; they are thus not discharged fromcare but receive regular checks with CT or MRI andthey receive either radiotherapy or chemotherapytreatment. Alternatively, the patient may receivepalliative care after the initial major treatment ifthere is recurrence or progression of the tumour(Table 26).Although a non-muscle-invasive tumour is notas likely to result in a serious life-threateningcondition, it has high recurrence rates. As discussedin Chapter 1, the recurrence rate of non-muscleinvasivedisease depends upon a number ofprognostic risk factors: stage, grade, size of thetumour and number of previous recurrences.Prognostic risk factors are essential to predictfuture courses of the tumour in terms of recurrenceand progression. Prognostic factors for recurrenceand progression have been investigated by severalclinical groups. The most frequent factor relatedto recurrence, in almost all series, has beenmultiplicity (Appendix 18, Table 55). Intravesicalinstillations have been defined as a protectivefactor. Kurth and colleagues 181 reported factorsaffecting recurrence and progression from thedata of two trials involving 576 patients. The trialsconsidered factors such as tumour size, grade, andrecurrence rate per year and concluded that themost significant prognostic factors for recurrencewere multiplicity, recurrence at 3 months, size ofthe tumour and site of involvement (Appendix18, Table 55). 20,181–195 Parmar and colleagues 191considered multiplicity and recurrence at 3 monthsas the main prognostic factors in recurrence. Thesetwo parameters provided the most predictiveinformation related to recurrence, and they wereindependent of the stage (Table 27). However, theMedical Research Council classification in Parmar’sstudy is only used to predict the risk of recurrence,not progression. 191Grade, associated CIS and stage are factorsglobally related to progression in the series thathave investigated prognostic factors (AppendixTABLE 26 Management of bladder cancerType of bladder cancerNon-muscleinvasiveMuscle invasiveLow riskIntermediate riskHigh riskInitial treatmentTURBT and one dose of mitomycinTURBT and one dose of mitomycinTURBT, one dose of mitomycin and BCG inductionThree cycles of chemotherapy and cystectomy or three cycles of chemotherapyand radiotherapy or palliative treatment78

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 418, Table 56). 20,181,183–185,187,192,196 Millán-Rodriguezand colleagues 187 developed three risk groupsbased on 1529 patients with primary non-muscleinvasivebladder cancer. The trial used recurrenceprognostic factors such as multiplicity, tumour sizeand CIS and progression prognostic factors such asgrade, CIS and multiplicity.Although different studies have analysed thefactors involved in recurrence and progression,there is no universally agreed prognostic risk groupclassification (Table 27). It is not possible to usethe risk stratification illustrated in Kurth’s study 181in the model because of the complexity of datarequirements for recurrence and progression. Therisk groups and their proportions will be definedlater in this chapter depending on the two studiesthat have the best data available for recurrence andprogression.Markov model structure for non-muscleinvasivediseaseAt the end of each risk group branch of nonmuscle-invasivedisease in the decision tree(Appendix 17, Figure 36) the patient will enterone of the following states of the Markov modelshown in Figure 32: (1) no tumour recurrence; (2)recurrence; (3) progression to muscle-invasivedisease; and (4) death. There are two diagnosticresults of non-tumour recurrence, i.e. true negativeand false negative, as well as true positive and falsepositive for tumour recurrence.The patients with a false-negative result in themodel will be followed using the follow-up strategyof non-tumour recurrence. The cycle lengthconsidered is 1 year, although the risk groups inthe care pathway will be followed at different timeperiods: 12 months for low risk, 6 months forintermediate risk and 3 months for high risk. Theabsorbing state is ‘death’, which can be reachedfrom any of the other states.Markov model for local muscle-invasivediseaseAt the end of each risk group branch of localmuscle-invasive disease in the decision tree(Appendix 17, Figure 38) the patient will enterTABLE 27 Studies of risk group classificationStudy Risk factors Proportion (%)Millán-Rodriguez 2000 187 Low risk TaG1, single T1G1 11.5Intermediate TaG2, multi T1G1 44.6riskHigh risk Multi T1G2 43.9TaG3, T1G3CISOosterlinck 2001 190 Low risk Single TaG1 and < 3 cm diameter NAIntermediate TaT1 excluding low and high risksNAriskHigh risk T1G3, CIS, multifocal or highly recurrent NAParmar 1989 191 Low risk Single tumour and no recurrence at first follow-up 60Intermediate Single tumour and no recurrence at first follow-up or 30riskmultiple tumour no recurrence at first follow-upHigh risk Multiple or highly recurrent 10Kurth 1995 181 Low risk G1 and no recurrence in 2 years 52.5G1, size (< 1.5 cm) and recurrence (< 3 cm) in 2 yearsG2, small size (< 1.5 cm) and no recurrence in 2 yearsIntermediate The others excluding low and high risks 40.7riskHigh risk G1, great size (> 3 cm) and > 3 recurrences in 2 years 6.7G2, great size (> 3 cm) and recurrence in 2 yearsG3NA, no details are available.79© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Assessment of cost-effectivenessLow riskNon-muscleinvasiveIntermediateriskHigh riskNo tumourrecurrenceDeathRecurrenceProgress to muscleinvasiveFIGURE 32 Markov model structure for non-muscle-invasive tumour.one of the following states of the Markov modelshown in Figure 33: (1) no tumour; (2) recurrence;(3) progression to metastases; and (4) death. Cyclelength will be the same as that of non-muscleinvasivedisease.Estimation of parameters usedin the modelParameters used in the decision tree and Markovmodels were calculated within the model orestimated from the systematic reviews of diagnosticperformance reported in Chapters 4 and 5 andthe epidemiology of bladder cancer reportedin Chapter 1, as well as other relevant costeffectivenessdata identified from the literature.The details of the data for the probabilities, costsand utilities used in the models are describedbelow.ProbabilitiesSensitivity and specificity of diagnostictestThe data on the sensitivity and specificity of eachdiagnostic test were taken from the systematicreview and are summarised in Table 28. For flexiblecystoscopy assessment there were no data availablefrom the systematic review. It is therefore assumedthat the accuracy of flexible cystoscopy used inthe models is the same as that of white light rigidcystoscopy. This assumption is relaxed in thesensitivity analysis in which the performance ofLocally muscleinvasiveNo tumourDeathRecurrenceMetastases80FIGURE 33 Markov model for local muscle-invasive follow-up.

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4TABLE 28 Data on diagnostic performanceDiagnosis Sensitivity 95% CI Specificity 95% CI SourceCSC 0.71 0.49 to 0.93 0.72 0.47 to 0.96 Systematic review based on WLCCTL 0.44 0.38 to 0.51 0.96 0.94 to 0.98 Systematic reviewNMP22 0.68 0.62 to 0.74 0.79 0.74 to 0.84 Systematic reviewImmunoCyt 0.84 0.79 to 0.91 0.75 0.68 to 0.83 Systematic reviewFISH 0.76 0.65 to 0.84 0.85 0.78 to 0.92 Systematic reviewPDD 0.92 0.8 to 1.0 0.57 0.36 to 0.79 Systematic reviewWLC 0.71 0.49 to 0.93 0.72 0.47 to 0.96 Systematic reviewCSC, flexible cystoscopy; CTL, cytology.flexible cystoscopy is increased by 5%, 10% andan extreme 20% compared with white light rigidcystoscopy.Prevalence rateThe prevalence rate was not derived from existingdata in the literature as the prevalence of bladdercancer varies considerably among subgroups withdifferent symptoms, from 1% to 20% (for men over50 years of age). 179 In the model base-case analysisit was assumed that the prevalence rate is 5% andin a sensitivity analysis a range of prevalence rateswas considered to identify those prevalence ratesfor which different diagnostic strategies may beconsidered worthwhile. This approach of repeatingthe analysis for different prevalence rates was feltto be more informative than defining prevalenceusing a wide uniform (i.e. uninformative)distribution.Proportions of types and theirsubgroups for bladder cancerThe proportions of the two main types of bladdercancer were assessed based on the literature andclinical opinions detailed in Chapter 1. Withreference to the available information presentedin the previous section and in Table 27, as wellas discussions with the clinical members of theresearch team, prognostic risk groups in nonmuscle-invasivedisease within this model havebeen categorised by using a combination ofMillán-Rodriguez and colleagues’ 187 classificationat initial diagnosis and Parmar and colleagues’ 191classifications at 3 months’ follow-up, i.e. lowrisk, intermediate risk and high risk. Theseclassifications are shown in Table 29, which alsoprovides details on the proportions of patients ineach risk group of non-muscle-invasive bladdercancer.Table 30 summarises the values of these proportionsused in the decision tree and Markov models.Recurrence, progression and mortalityof non-muscle-invasive diseaseTable 31 shows the probabilities of recurrence,progression and mortality for the three riskgroups of non-muscle-invasive disease used in themodel for a 20-year time horizon. As referred toabove, the first 5-year probabilities of recurrence,progression and mortality caused by cancer ofthe three risk groups used in the model werecalculated from the study by Millán-Rodriguez andcolleagues. 187 The following 15-year probabilitiesof recurrence, progression and mortality causedby cancer in these groups were estimated byusing mean values of relevant data of the last 3years in the 5-year data available in the study byMillán-Rodriguez and colleagues. 187 This was aretrospective cohort study of 1529 patients withprimary non-muscle-invasive bladder cancerin Spain in the years 1968–96. Of the patientstreated with TURBT and random biopsy, halfwere treated using additional BCG and one-thirdusing additional intravesical instillation (mainlymitomycin C, thiotepa and doxorubicin). However,the characteristics of the patients, such as genderand mean age, were not reported, and the followupwas less than 5 years. 18781© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Assessment of cost-effectivenessTABLE 29 Risk group stratificationRisk groupsLow: TaG1, singleT1G1Intermediate: TaG2,multi T1G1, singleT1G2High: TaG3, T1G3,CIS, multi T1G2Subgroups (cancer at diagnosis)Factors defined in follow-up at 3months Proportion (%)Group 1: single TaG1, single T1G1 No tumour recurrence 10Group 2a: single TaG1, single T1G1 Tumour recurrence aGroup 2b: multi TaG1No tumour recurrenceGroup 3: multi TaG1Tumour recurrence aGroup 1: single TaG2, single T1G2 No tumour recurrence 45Group 2a: single TaG2, single T1G2 Tumour recurrence aGroup 2b: multi TaG2, multi T1G1 No tumour recurrenceGroup 3: multi TaG2, multi T1G1 Tumour recurrence aTumour recurrence or not 45a If TaG3, T1G3, CIS, multi T1G2 recurrence, then joins high-risk treatment pathway.TABLE 30 Proportions of types and their subgroups for bladder cancerType of bladder cancer Proportion Subgroups of bladder cancer considered ProportionNon-muscle invasive 75% Low risk 10%Intermediate risk 45%High risk 45%Muscle invasive 25% Local muscle invasive 75%Metastases 25%Recurrence, progression and mortalityof muscle-invasive diseaseWhen patients move into the Markov modelfor muscle-invasive disease, the model requiresestimates of the annual rates of recurrence,progression and mortality caused by cancer.The probabilities of recurrence, progressionand mortality of muscle-invasive disease andmetastases used in the model for 20 years arepresented in Table 32. The first 5-year probabilitiesof recurrence, progression and mortality causedby local muscle-invasive disease used in the modelwere obtained from a retrospective cohort study inCanada by Stein and colleagues 197 in which a cohortof 1054 patients with muscle-invasive bladdercancer were treated by radical cystectomy between1971 and 1997. The mean age of the patients was66 years, 80% of the patients were male 197 anddata were available for 10 years of follow-up. Thelast 10-year probabilities used in the model areassumed to be the same as the data reported forbetween 5 and 10 years in the study by Stein andcolleagues. The last column of Table 32 presents theprobabilities of mortality for metastases providedby von der Maase and colleagues 198 and there aredata available for 5 years of follow-up. The last5-year probabilities used in the model are assumedto be the same as rates reported for between 3and 5 years in von der Maase and colleagues. 198This RCT investigated the long-term survival ofpatients with metastatic bladder cancer treated withchemotherapy in Denmark. Of the 405 patients,137 had locally advanced disease and 268 hadmetastatic disease. The median survival time was8.3 months.All-cause mwortality rates in the UKAs patients progress through the model over time,values of annual rates of age-specific general orall-cause mortality are required. These were takenfrom the published UK life tables for the years2004–6. 199 As discussed in Chapter 1, Cancer82

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4TABLE 31 Probabilities of recurrence, progression and mortality in non-muscle-invasive bladder cancerTime(years)Recurrence (%) Progression (%) Mortality caused by cancer (%)Low Intermediate High Low Intermediate High Low Intermediate High3 months 2 4 9.4 0 0.2 1.3 0 0 01 15 26 39 0 0.4 8 0 0.4 12 10 13 11 0 0.8 5 0 0 33 5 6 6 0 0.6 3 0 0.3 14 8 5 2 0 0.8 1 0 0.3 25 7 3 3 0 1 2 0 0 26 7 5 4 0 0.8 2 0 0.2 27 7 5 4 0 0.8 2 0 0.2 28 7 5 4 0 0.8 2 0 0.2 29 7 5 4 0 0.8 2 0 0.2 210 7 5 4 0 0.8 2 0 0.2 211 7 5 4 0 0.8 2 0 0.2 212 7 5 4 0 0.8 2 0 0.2 213 7 5 4 0 0.8 2 0 0.2 214 7 5 4 0 0.8 2 0 0.2 215 7 5 4 0 0.8 2 0 0.2 216 7 5 4 0 0.8 2 0 0.2 217 7 5 4 0 0.8 2 0 0.2 218 7 5 4 0 0.8 2 0 0.2 219 7 5 4 0 0.8 2 0 0.2 220 7 3 4 0 0.8 2 0 0.2 2Research UK reported that 70% of all primarybladder cancer affects men and therefore the allcausemortality for the model cohort was weightedto reflect this (Figure 34). Further data related tothe rate of all-cause mortality are shown in Table 57in Appendix 18.Other probabilitiesMortality rates of WLC/PDD and TURBTWhite light rigid cystoscopy (WLC), PDD andTURBT are invasive procedures. As with allsurgical procedures requiring general anaesthetic,death due to complications in the perioperativeperiod is a potential risk. There are no availabledata on mortality rates associated with WLC orPDD. The probability of death during WLC andPDD in Table 33 was therefore obtained from astudy by Farrow and collegues, 200 which examined108,878 anaesthetic cases in Cardiff between1972 and 1977. The probability of death duringTURBT in Table 33 was obtained from Kondas andcolleagues, 201 which evaluated 1250 TURBT casesin Cardiff during 18 years.© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.Relative risk for progression comparing notreatment (false negative) with treatment (truepositive)As some patients who have bladder cancer shownegative results during the initial diagnosisor follow-up, it was believed that the risk ofprogression in the case of a false negative withoutrelevant treatment was higher than that of a truepositive with treatment. However, there are no dataavailable in relation to false-negative diagnoses.Although there are some studies investigatingdisease-free survival or survival for different typesof drug treatment as an adjunct to initial treatment(TURBT) for bladder cancer, there is no identifiedstudy that compares survival with and withoutTURBT. Using information from the Millán-Rodriguez and colleagues’ study 187 it was assumedthat the base-case RR for progression comparingno treatment (TURBT) with treatment (TURBT)was 2.56, that is the RR compared TURBT plusBCG with TURBT alone. The uncertainty aroundthis value was tested as part of the sensitivityanalysis.83

Assessment of cost-effectivenessTABLE 32 Probabilities of recurrence, progression and mortality in muscle-invasive bladder cancerLocal muscle-invasive disease after cystectomyMetastasesTime (years) Recurrence (%) Progression (%) Mortality (%) Mortality (%)3 months 0 6.25 3 10.51 0 25 12 422 0 13 11 803 0 8 9 504 0 4 8 505 0 4 8 506 0 4 7 507 0 4 6 508 0 4 5 509 0 4 5 5010 0 4 5 5011 0 4 512 0 4 513 0 4 514 0 4 515 0 4 516 0 4 517 0 4 518 0 4 519 0 4 520 0 4 5Survival probability1.00.90.80.70.60.50.40.30.20.10.00 10 20 30 40 50 60 70 80 90 100AgeFIGURE 34 Kaplan–Meier plot for sex- and age-adjusted survival (30% female, 70% male) in the UK.84

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Relative risks for recurrence and progressioncomparing PDD with WLC treatmentOne of the issues that could be considered in themodel is whether the recurrence and progressionrates of non-muscle-invasive disease differ basedon the type of intervention used in the treatment(PDD or WLC). Although there is some evidencein Chapter 4 that PDD may reduce recurrenceand progression for non-muscle-invasive diseasecompared with WLC, there are no reliable datarelated to recurrence and progression of nonmuscle-invasivebladder cancer following PDDor WLC in primary diagnosis. It was thereforeassumed that recurrence and progression ratesare not different between PDD and WLC so thatthe base-case RR for recurrence and progressioncomparing PDD and WLC is 1. This assumptionwas tested as part of the sensitivity analysis.Probability of detecting missed bladder cancer afterfalse-negative resultsThere is no evidence to suggest when patients whohave false-negative results should be detected.Therefore assumptions were made about whensuch patients were identified. The probabilities ofdetecting false-negative cases are described in Table33.CostsTable 34 shows the cost estimates for the testsand investigations used within the model. Thecosts of flexible cystoscopy, WLC or WLC-assistedTURBT were identified from 2006 NHS referencecosts. 202 The cost of flexible cystoscopy was basedon the NHS reference cost with HealthcareResource Group (HRG) (day case) code L21‘Bladder cancer endoscopic procedure withoutcomplications (cc)’. The cost of WLC was basedon the NHS reference cost with HRG (electiveinpatient) code LB15C ‘Bladder minor procedure19 years and over without cc’. The day unit costof WLC-assisted TURBT was based on the NHSreference cost with HRG (elective inpatient) codeL21 ‘Bladder intermediate endoscopic procedurewithout cc’. Based on the 2006 report by Karl StorzEndoscopy (UK), the cost of WLC-assisted TURBTis calculated by multiplying the cost per day by 2days. [Karl Storz Endoscopy (UK), 2006, personalcommunication]. Also reported in Table 34 are thecosts of PDD. Compared with WLC, PDD incursthe following additional costs:• extra equipment: photosensitiser (HAL, ALA),colour CCD camera (on chip integration),xenon lamp, fluid light cable• extra personnel involved: unlike WLC, PDDrequires the instillation of a photosensitiservia a urethral catheter prior to TURBT; this isusually performed by a nurse on the ward• procedure time: on the ward, catheterisationand instillation of photosensitiser and thenremoval of catheter takes about 15 minutes; intheatres, fluorescence-guided TURBT takes anextra 10 minutes compared with conventionalwhite light TURBT alone.The additional cost of extra equipment, personneland time of PDD were obtained from a businessreport prepared by Karl Storz (UK) [Karl StorzEndoscopy (UK), 2006, personal communication](Table 35). It was assumed that the lifespan of PDDequipment is 5 years, a 3.5% discount rate is usedin equivalent annual cost and the average numberof PDD tests per year is 100.The costs associated with the additional resourcesare shown in Table 36 and these costs were addedto the costs of WLC to obtain the costs of PDD andPDD-assisted TURBT.TABLE 33 Other probabilitiesOther probabilities Value SourceMortality rate of WLC/PDD 0.5% Farrow 1982 200Mortality rate of TURBT 0.8% Kondas 1992 201False negatives: probability detected in first 3 months 50% AssumptionRelative risk for progression (no treatment vs treatment) 2.56 Millán-Rodriguez 2000 187Relative risk for recurrence (PDD vs WLC) 1 AssumptionRelative risk for progression (PDD vs WLC) 1 AssumptionFalse negatives: probability detected in first year 50% AssumptionFalse negatives: probability detected in second year 75% AssumptionFalse negatives: probability detected after second year 100% Assumption85© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Assessment of cost-effectivenessTABLE 34 Cost of diagnostic tests and initial treatments for bladder cancerParameter Base case (£) Range Unit SourcePDD 1371 1136 to 1758 Procedure Health Care FinancialWLC 937 702 to 1324 Procedure NHS reference costs 202CSC 441 362 to 680 Session NHS reference costs 202Cytology 92.37 Uniform distribution Session NHS reference costs 202NMP22 39.30 25 to 54.8 Test MediChecks.comImmunoCyt 54.8 Uniform distribution Session NHS reference costs 202FISH 54.8 40 to 60 Test NHS reference costs 202PDD-assisted TURBT 2436 2006 to 2994 Procedure Health Care FinancialWLC-assisted TURBT 2002 1572 to 2560 Procedure NHS reference costs 202CT scan 325 Uniform distribution Procedure Rodgers 2006 179CSC, flexible cystoscopy.TABLE 35 Estimated additional costs for extra capital resource of PDDAdditional capital resourceCostTotal cost of the extra equipment for PDD £17,950Lifespan of the equipment (years) 5Average number of PDD tests per year 1003.5% discount rate for 5 years 0.2215Equivalent annual cost £3976Additional cost per test £40Cost of hexyl-5-aminolaevulinic acid per test £286Annual service and maintenance costs (after year 1) £1795Cost of service and maintenance per patient £18Total average cost per test £344TABLE 36 Estimated additional costs for incorporating the PDD procedureAdditional procedureAdditional costExtra nurse time for catheterising patients and instillation of 5-ALA £40Extra staffing cost (operation) £35Additional equipment of PDD £344Consumables (catheter, etc.) £15Total £43486The states related to ‘true negative’ and ‘falsenegative’ only incur diagnostic costs. However, thestates for ‘true positive’ and ‘false positive’ incurboth diagnostic and relevant treatment costs. Forexample, for strategy CSC_WLC, the costs of ‘truepositive of low risk’ and ‘false positive of low risk’are equal to cost_CSC. The costs of ‘true negativeof low risk’ and ‘false negative of low risk’ are equalto cost_CSC + cost_TURBT. For muscle-invasivedisease relevant diagnostic and treatment costswere also considered.

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4The cost of NMP22 was based on the marketingprice in the UK. 203 As the costs of ImmunoCyt andFISH are not available in the UK market, thesecosts were calculated from a systematic reviewconducted for NICE 179 as well as from 2005 NHSreference costs 202 with HRG code L13 ‘Minorpathology test’. The cost of cytology was estimatedusing HRG code L14 ‘Intermediate pathologytest’ 179 and the cost of a CT scan was estimated byusing data from the same source. 179Table 37 reports the costs of treatments forbladder cancer. The cost of cystectomy was basedon 2006 NHS reference costs with HRG codeLB389B ‘Cystectomy with urinary diversion andreconstruction without cc’. The unit day cost ofpalliative treatment was also obtained from NHSreference costs with HRG code SD01A ‘Inpatientspecialist palliative care 19 years and over’.Following consultation with clinical experts, anassumption was made that the palliative treatmentrequires a range of 3–6 months. The cost ofpalliative treatment was estimated by multiplyingthe unit cost per day by 135 days. This figure isuncertain as it would of course depend upon thetype of care necessary. However, the proportion ofpatients likely to need this care is relatively smalland the likely differences between strategies willalso be small.The unit cost of radical radiotherapy was obtainedfrom Aberdeen Royal Infirmary (Dr GhulamNabi, University of Aberdeen, May 2008, personalcommunication). Radical radiotherapy requiresfrom 30 to 40 sessions. The cost of radiotherapywas calculated by multiplying the unit cost by 35sessions. The costs of the three drug treatments –mitomycin, BCG and cisplatin – were derived fromthe British National Formulary (http://bnf.org).Discount rateDiscount rates used for costs and outcomeswere those recommended in the recent NICEguideline 204 on the conduct of technologyassessment reviews. Annual discount rates of 3.5%with a range from 0% to 6% were used in themodel.Estimation of total cost of strategiesThe total cost for each strategy was determinedusing recursive costing in the decision tree and theMarkov model. At the end point in the decisiontree model this is achieved by setting the costvariable as 0 at the root node. As the tree expandsfrom left to right, the ‘cost’ variable is modified byadding new cost variables to the variable ‘cost’. Inthis way, the value of ‘cost’ at each terminal nodeis unique to the path from the root node to thatterminal node. In the example strategy being used,flexible cystoscopy followed by WLC, the value of‘cost’ at the ‘true-positive’ terminal node would bethe costs of flexible cystoscopy and WLC and theadditional treatment cost depending on the type ofbladder cancer.Discounted costs are considered in the Markovmodel to estimate the cost for each diagnosticstrategy by using the following formulation:Cost = cost discount ratestrategy ∑ / ( 1+)cyclecycleTABLE 37 Cost of treatment and management of bladder cancerParameter Base case (£) Range Quantity Unit SourceMitomycin 73.88 Uniform 40 mg Cycle British National FormularydistributionBCG 89 Uniform 12.5 mg Cycle British National FormularydistributionCystectomy (w/o cc) 6856 3656 to 8437 Procedure NHS reference costs 202Chemotherapy (cisplatin) 50.22 25.37 to 100 Cycle British National FormularyRadical radiotherapy 1050 900 to 1200 35 (30–40) £30/day Aberdeen Royal InfirmaryPalliative treatment(outpatient)12,825 8550 to17,100£95/day NHS reference costs 202Discount 3.5% 0% to 6% NICE guideline 204w/o cc, without complications.87© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Assessment of cost-effectivenessDistribution of parametersFor probabilities of recurrence, progression andmortality of bladder cancer and all-cause mortalityrate, no distribution was assigned, as the numberof observations or studies used to calculate the riskwas very large. The estimates of sensitivity andspecificity of the three biomarker tests and cytologywere assigned normal distributions, which appearto fit the data that have small and symmetricranges. The estimates for the performance offlexible cystoscopy, WLC and PDD were assignedbeta distributions, which are more flexible todeal with data that have large and skewed ranges.Diagnoses and treatment costs were assigned lognormaldistributions as this distribution appearedto best fit the data that have skewed or symmetricranges.Quality of life measuresTo conduct a cost–utility analysis, quality of life(QoL) (utilities) data are required. The bestestimates of QoL (utilities) data for a UK settingmay be provided by using generic measures suchas EQ-5D or SF-6D (which might be derived fromresponses to the SF-36 or SF-12). A structuredliterature search was conducted in EMBASE,MEDLINE and other relevant databases usingthe key words related to urological cancer, EQ-5Dand SF-36 (Appendix 1). However, no QoL datawere identified relating to bladder cancer. Theonly available QoL data were for other urologicalcancers. After discussions with clinical expertsinvolved in this study it was decided not to useQoL estimates for other urological cancers as aproxy as these values were not considered to begeneralisable to the population who have bladdercancer, although as reported later sensitivityanalysis was conducted that explored the impact ofusing these data.Data analysisCost-effectiveness analysisThe base-case analysis was based on the costsand outcomes for a hypothetical cohort of 1000people with a mean age of 67 years reported inthe systematic review in Chapter 4. The basecasemodel analysis was run for 5% prevalencerates and a 20-year time horizon. Two differentmeasures of incremental cost-effectiveness havebeen considered as they provide slightly differentinformation. These measures are the incrementalcost per true positive case detected and incrementalcost per life-year gained. The cases of true positivesmight be considered to be the key clinical outcometo reflect the diagnostic performance and life-yearsare a natural outcome to reflect survival.The incremental cost-effectiveness is presentedboth with and without dominated and extendedlydominated options. For the estimation ofincremental cost per life-year gained the resultsare presented as cost-effectiveness scatter plots andcost-effectiveness acceptability curves (CEACs).CEACs illustrate the likelihood that the strategyis cost-effective at various threshold values forsociety’s willingness to pay for an additional lifeyear.Probabilistic sensitivity analysis was basedmainly on the non-dominated strategies in thebase-case model as changes in the estimates ofparameters in these particular strategies are morelikely to change the conclusions.Cost–consequence analysisThe cost-effectiveness analysis results werepresented as true positive cases detected and lifeyears.Further information can be obtained byconsidering the different outcome of diagnosticperformance and longer-term effectiveness withinthe model for each strategy included in this study.The diagnostic performance of each strategy isreported in terms of false negative, false positive,true negative, correct diagnosis and incorrectdiagnosis. Here, data along with information onlife expectancy and cost can be presented in theform of a cost–consequence analysis. As such thesedata can be useful to aid in the interpretationof cost-effectiveness analyses and, had one beenpossible as part of the base-case analysis, a costutilityanalysis as they help to identify what factorsmight be drivers of the results.Sensitivity analysisSensitivity analyses were carried out to exploreuncertainties within the model. Sensitivity analysesconcentrated on various assumptions made aboutestimates of main parameters used in the basecasemodel. As mentioned above the results of thesensitivity analyses focused on the non-dominatedstrategies in the base-case model. A cost–consequence analysis can be used to highlight thechoices and trade-offs that can be made betweenoutcomes.Prevalence rates of patients who havesymptoms of bladder cancerAlthough considerable efforts were made toidentify estimates for prevalence rates for patientswho have symptoms of bladder cancer, no reliabledata were available. In the base-case analysis a88

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4prevalence rate of 5% was used. Existing data inthe literature suggest that prevalence rates rangefrom 1% to 20%. Sensitivity analysis was performedto explore the effects of a decrease to 1% andincreases to 10% and 20%. The same distributionof parameters adopted in the base-case analysis wasused.Relative risk of progression comparingno treatment (false negative) andtreatment (true positive)As mentioned earlier there was little informationavailable to investigate the risk of progression ofno treatment for patients who have bladder cancerwhen they have negative results in the initialdiagnosis. Bladder cancer missed in the initialdiagnosis and at follow-up would not be treatedand would subsequently have a higher risk ofprogression and mortality. The base-case analysisassumed that the RR of no treatment (TURBT)compared with treatment would be 2.56 basedon the Millán-Rodriguez and colleagues’ study. 187A range of this RR was considered to investigatethose values for which diagnostic strategies may beconsidered worthwhile. Based on available evidenceon the RR for progression comparing TURBTwith TURBT plus BCG or other drugs, a sensitivityanalysis was performed with the assumption thatthe RR for progression comparing TURBT with noTURBT decreased to 1.Relative risks of recurrence andprogression comparing PDD with WLCThere are no reliable data on recurrence andprogression when PDD is used for initial diagnosisand follow-up, although PDD is likely to reducerecurrence and progression compared with WLCas described in Chapter 4. It was assumed in thebase-case model that the RRs of recurrence andprogression comparing PDD with WLC would be1, i.e. any gains from the use of PDD would flowfrom improvements in diagnostic performanceas measured by sensitivity and specificity alone,as opposed to gains that might arise from a morecomplete removal of the cancer facilitated by theincreased information provided by PDD. Resultsin Chapter 4 suggested that the RRs of recurrenceand progression comparing PDD with WLC were0.64 and 0.56 and these values were used in thesensitivity analysis.Sensitivity and specificity of flexiblecystoscopyThere were no data related to the sensitivity andspecificity of flexible cystoscopy, although it islikely that the performance of flexible cystoscopy© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.could be better than that of WLC. The assumptionwas made in the base-case analysis that theperformance of flexible cystoscopy would bethe same as that of WLC. Expert opinion (TRLeyston Griffiths, University of Leicester, July2008, personal communication) suggested thatthe performance of flexible cystoscopy is betterthan that of WLC; sensitivity analysis was thereforeperformed assuming that both sensitivity andspecificity of flexible cystoscopy are increased from5% to 25% compared with WLC.Proportion of risk groups for nonmuscle-invasivebladder cancerThe risk groups used in the model were definedby combining two classifications based on the bestavailable data. There were large differences inthe proportions for risk groups in the two studies.The base case assumed that the proportion of riskwould be the same as in the Millán-Rodriguez andcolleagues’ study, 187 in which the proportion ofthe high-risk group is much higher than that ofthe low-risk group. As mentioned in Chapter 1 itis likely that the proportion of the low-risk groupin non-muscle-invasive disease is the same as thatin the study by Parmar and colleagues. 191 Thus,it was assumed in the sensitivity analysis that theproportion of the high-risk group decreased from30% in the base-case analysis to 10% and thatthe proportion of the low-risk group increasedfrom 10% in the base-case analysis to 30%. Thedistributions of parameters were the same as thoseused in the base case.Starting age and 10-year time horizonAs mentioned in Chapter 1 the incidence andmortality rate of bladder cancer are likely toincrease as age increases. The base-case analysiswas carried out on the assumption that the startingage of the cohort would be 67 years, based on theresults from the systematic review, and considered a20-year time horizon with constant mortality ratesof bladder cancer except for the first 5 years. Thesensitivity analysis used the reported mean age ofbladder cancer patients in the UK of 71 years. Theprevalence and mortality rate of bladder cancerassociated with age may imply that the most costeffectivestrategy in the base case may no longer beconsidered to be cost-effective.Annual discount rateAs recommended in the NICE guidelines, anannual discount rate of 3.5% for cost and outcomeswas used in the base-case model. A range from0% to 6% for discount rate was considered in thesensitivity analysis.89

Assessment of cost-effectivenessFollow-up diagnostic strategiesWhite light rigid cystoscopy was considered asthe second-line test in follow-up for each strategyin the base-case model as it is commonly used tofollow bladder cancer in the UK if the result of thefirst test in follow-up is positive. Sensitivity analysiswas performed to investigate whether alternativestrategies associated with PDD in follow-up maybe more cost-effective than those involving WLC,although PDD is more expensive than WLC.Quality of life measuresAs addressed in the previous section cost–utilityanalysis was not conducted in the base case.Sensitivity analysis was performed using the QoLdata from other urological cancers to producequality-adjusted life-years (QALYs). The utilityvalues identified for urological cancers are includedin Table 38. A prediagnosis utility value of 0.78 wasidentified and the rest of the values were based ona reduction in utility for undergoing the differenttests and treatments.Subgroup analysisDepending on data availability it was intended thatsubgroup analysis would be performed on:• type of tumour detected, e.g. CIS, low risk andhigh risk• tumour recurrence at the first 3-monthcystoscopic examination following TURBT• diagnostic performance of the different PDDphotosensitising agents.ResultsDeterministic and probabilisticresultsThe cost-effectiveness analysis aggregates thediagnostic performance and the time spent in thevarious health states of the model. As describedpreviously cost–utility analysis was not performedbecause QoL data suitable for incorporation intothe economic model were not available.TABLE 38 Utility valuesUtility and disutilityAssumption of reductionin utility Value Range SourcePrediagnosis NA a 0.78 0.52 to 1.0 UK EQ-5DCSC –0 0.78 0.518 to 1.0 Kulkarni 2007 205CTL –0 0.78 0.52 to 1.0 AssumptionNMP22 –0 0.78 0.52 to 1.0 AssumptionImmunoCyt –0 0.78 0.52 to 1.0 AssumptionFISH –0 0.78 0.52 to 1.0 AssumptionWLC –0.05 0.73 0.66 to 0.73 Kulkarni 2007 205PDD –0.05 0.73 0.66 to 0.73 Kulkarni 2007 205TURBT –0.05 0.73 0.66 to 0.73 Kulkarni 2007 205BCG –0.016 0.764 0.534 to 0.764 Kulkarni 2007 205Cystectomy (alone) NA b 0.624 0.39 to 0.78 Kulkarni 2007 205Chemotherapy –0.28 0.60 0.08 to 0.62 Kulkarni 2007 205Radiotherapy –0.13 0.65 0.49 to 0.65 Pickard 2007 206Non-muscle-invasive –0 0.78 0.24 to 0.73 Kulkarni 2007 205Muscle-invasive –0 0.78 0.52 to 1.0 UK EQ-5DMetastases with palliativetreatment–0.29 0.49 0.518 to 1.0 Kulkarni 2007 205CSC, flexible cystoscopy; NA, not applicable.a Not applicable as this is the starting value from which reductions are made.b Not applicable data based on that from Kulkarni and colleagues 2007. 20590

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Deterministic resultsThe cost-effectiveness of the 26 strategies for initialdiagnosis and follow-up were considered over a 20-year time horizon.Base case: diagnostic performance andlife-years and costs per patientTable 39 shows the results for a hypothetical cohortof 1000 patients. The table reports performanceof the strategies, from the least to the most costly.For each strategy the diagnostic performance of thestrategy and the average cost and life expectancyover a 20-year time horizon are shown. It isimportant to remember when interpreting thesedata that in the base-case analysis the prevalence ofdisease is 5% (i.e. 50 people out of the 1000 in thecohort have bladder cancer).Of the strategies shown, strategy 26, flexiblecystoscopy and ImmunoCyt followed by PDD ininitial diagnosis and flexible cystoscopy followed byWLC in follow-up [CSC_IMM_PDD (CSC_WLC)],has the best performance in terms of the highestnumber of true positives and lowest number offalse negatives and the highest number of life-yearsbut it also has the worst performance in terms ofthe highest number of false positives (n = 188), thelowest number of true negatives and the highestcost. Strategy 1, CTL_WLC (CTL_WLC), reportsthe lowest numbers of true positives and falsepositives and life-years saved and the highest valuesfor true negatives and false negatives.Cost–consequence analysisThe results presented in Table 39 can be usedto consider the trade-offs between the differenttreatment strategies and this can be furtherillustrated using the data presented in Table 40.Table 40 reports the strategies that perform thebest in terms of the different outcome measuresconsidered. The results for all strategies arereported in Appendix 19 (Tables 58 and 59). Forexample, CSC_IMM_PDD (CSC_WLC) is the bestperformingstrategy in terms of having the lowestfalse-negative and the highest true-positive ratesand longest survival. However, it is associated withthe highest rates of false positives and the lowestrates of true negatives.This table and Table 39 illustrate the trade-offs thatexists between those strategies that can correctlyidentify those without disease but will result in allof the harms from an incorrect diagnosis comparedwith those strategies that are better able to identifydisease if it is present but also result in additionalanxiety and cost for those incorrectly initiallydiagnosed as positive.Cost-effectiveness analysisIncremental cost per true positive casedetectedThe cost-effectiveness results for diagnosticperformance are presented in Table 41 usingincremental cost per true positive detected. Interms of mean true positive cases and costs,most of the strategies associated with flexiblecystoscopy or WLC in the initial diagnosis [exceptfor CTL_WLC (CTL_WLC) and FISH_WLC(FISH_WLC)] are dominated by those that involvePDD or biomarkers and can be eliminated becausethey are less effective and more costly than thenon-dominated strategies. The lower part of thetable reports the incremental cost-effectivenessratios (ICERs) when dominated and extendedlydominated strategies are omitted.The results in Table 41 show that strategy 26(CSC_IMM_PDD) has the highest number of truepositive cases detected (n = 44) and is the mostcostly strategy (£2370) per patient. Strategy 1(CTL_WLC) has the lowest cost per patient (£1043)and produces the least number of true positives(n = 16). It is also highlighted in the table that totalcost increases when moving from WLC to PDD andthe number of cases detected also increases whenPDD is used.Incremental cost per life-yearThe base-case analysis was also presented in termsof incremental cost per life-year (Table 42). Theresults presented for life-years are similar to thosepresented in Table 41. As can be seen from Table42 many strategies are dominated, that is theyprovide no more or even less benefits at the sameor increased cost. Further strategies are extendedlydominated, that is providing a mix of a lower costbut less effective strategy and a higher cost butmore effective strategy would be more efficient.The strategy of FISH_WLC (FISH_WLC) isextendedly dominated by the strategy of CTL_PDD(CTL_WLC) and it can be eliminated as its ICERis greater than that of FISH_PDD (FISH_WLC)as well as CSC_IMM_PDD. Furthermore, evenfor those strategies that are not dominated orextendedly dominated the incremental cost perlife-year gained might be higher than societyis willing to pay. Reference values for society’swillingness to pay for a life-year are not availablebut given that people will be in less than full healthit is likely that the incremental cost per QALY91© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Assessment of cost-effectivenessTABLE 39 Results of the deterministic model for the 20-year time horizonStrategyDiagnostic performanceAveragelimitationoutcomeFirst line tests(second linetests)1 CTL_WLC(CTL_WLC)2 CTL_PDD (CTL_WLC)3 FISH_WLC(FISH_WLC)4 FISH_PDD(FISH_WLC)5 NMP22_WLC(NMP22_WLC)6 NMP22_PDD(NMP22_WLC)7 IMM_WLC(IMM_WLC)8 IMM_PDD (IMM_WLC)9 CSC_CTL_WLC(CTL_WLC)10 CSC_FISH_WLC(FISH_WLC)11 CSC_NMP22_WLC (NMP22_WLC)12 CSC_CTL_PDD(CTL_WLC)13 CSC_WLC(CSC_WLC)14 CSC_IMM_WLC(IMM_WLC)15 CSC_CTL_WLC(CSC_WLC)16 CSC_FISH_WLC(CSC_WLC)17 CSC_FISH_PDD(FISH_WLC)18 CSC_NMP22_WLC (CSC_WLC)19 CSC_PDD (CSC_WLC)20 CSC_NMP22_PDD (NMP22_WLC)TruepositiveTruenegativeFalsepositiveFalsenegativeCorrectdiagnosisIncorrectdiagnosisLifeyearsCost16 939 11 34 955 45 11.59 £104320 934 16 30 954 46 11.6 £109427 910 40 23 937 63 11.62 £117135 889 61 15 924 76 11.64 £123524 894 56 26 918 82 11.61 £124231 864 86 19 895 105 11.62 £132130 884 67 20 913 87 11.63 £134539 848 102 11 887 113 11.65 £145830 868 82 20 898 102 11.62 £166233 847 103 17 880 120 11.63 £180732 835 115 18 867 133 11.62 £185139 824 126 11 863 137 11.65 £185925 876 75 25 901 99 11.6 £192034 828 122 16 862 138 11.63 £194130 868 82 20 898 102 11.62 £199733 847 103 17 880 120 11.66 £200543 792 158 7 835 165 11.63 £204232 835 115 18 867 133 11.62 £207033 836 114 17 869 131 11.63 £208242 774 176 8 816 184 11.65 £208992

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4TABLE 39 Results of the deterministic model for the 20-year time horizon (continued)StrategyDiagnostic performanceAveragelimitationoutcomeFirst line tests(second linetests)21 CSC_IMM_WLC(CSC_WLC)22 CSC_CTL_PDD(CSC_WLC)23 CSC_IMM_PDD(IMM_WLC)24 CSC_FISH_PDD(CSC_WLC)25 CSC_NMP22_PDD (CSC_WLC)26 CSC_IMM_PDD(CSC_WLC)TruepositiveTruenegativeFalsepositiveFalsenegativeCorrectdiagnosisIncorrectdiagnosisLifeyearsCost34 828 122 16 862 138 11.63 £210539 818 132 11 857 143 11.64 £214544 762 188 6 806 194 11.66 £219543 792 158 7 835 165 11.66 £227042 774 176 8 816 184 11.65 £231844 762 188 6 806 194 11.66 £2370would be greater than £20,000 for all strategiesapart from 2, 3 and 4. The incremental cost perQALY for strategy 8 may be greater than £20,000but less than £30,000 as long as the average annualQoL score is 0.65.Probabilistic resultsThe cost-effectiveness point estimates donot provide any information on uncertaintysurrounding the model parameters. The resultsof the probabilistic analysis revealed the level ofuncertainty concerning results as illustrated in theCEACs in Figure 35.As can be seen in Figure 35 none of the eightstrategies considered is likely to be cost-effectivemore than 50% of the time when society is willingto pay relatively little for an additional life-yearexcept for strategy 1 [CTL_WLC (CTL-WLC)].Nevertheless, there are four strategies that are eachassociated with an approximately 20% chance ofbeing considered cost-effective over much of therange of willingness to pay values considered. Itis notable that three of the four strategies involvethe use of biomarkers for diagnosis and follow-up,while the fourth uses cytology.As mentioned in the methods section of thischapter, the cost-effectiveness estimates for thosestrategies that involve more than one test as partof the initial diagnosis may be underestimated.Adding in these potential extra costs hadvirtually no effect on the point estimates of costeffectivenessor on the likelihood that a particularstrategy would be likely to be considered costeffective.Sensitivity analysis and subgroupanalysisChanging prevalence rates in patientswho have symptoms of bladder cancerAs prevalence rates increase, people with suspectedbladder cancer have more positive results andthe costs and outcomes associated with diagnosticperformance for each strategy are increased.However, the outcomes associated with longtermsurvival may be decreased, because fewerpeople within the cohort are disease free. Table43 describes the results of the sensitivity analysisfor changes in the prevalence rate. The nondominatedor non-extendedly dominated strategiesare the same as in the base-case analysis and areexcluded from the table. At low probabilities ofdisease (i.e. 1%) it is likely that the least costlystrategy, strategy 1 [CTL_WLC (CTL_WLC)],is likely to be cost-effective. The probability ofIMM_PDD (IMM_WLC), FISH_PDD (FISH_WLC)and CSC_FISH_PDD (FISH_WLC) being alsoconsidered as cost-effective strategies at differentthresholds of society’s willingness to pay for anadditional life-year in the base case did not vary93© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Assessment of cost-effectiveness94TABLE 40 Ranking by diagnostic and life-year performance and costIncorrectdiagnosis Life-years CostCorrectdiagnosisRanking True negative True positive False positive False negative1 CTL_WLC CSC_IMM_PDD CTL_WLC CSC_IMM_PDD CTL_WLC CTL_WLC CSC_IMM_PDD CTL_WLC(CTL_WLC) (CSC_WLC) (CTL_WLC) (CSC_WLC) (CTL_WLC) (CTL_WLC) (CSC_WLC) (CTL_WLC)2 CTL_PDD CSC_IMM_PDD CTL_PDD CSC_IMM_PDD CTL_PDD CTL_PDD CSC_IMM_PDD CTL_PDD(CTL_WLC) (IMM_WLC) (CTL_WLC) (IMM_WLC) (CTL_WLC) (CTL_WLC) (IMM_WLC) (CTL_WLC)3 FISH_WLC CSC_FISH_PDD FISH_WLC CSC_FISH_PDD FISH_WLC FISH_WLC CSC_FISH_PDD FISH_WLC(FISH_WLC) (FISH_WLC) (FISH_WLC) (FISH_WLC) (FISH_WLC) (FISH_WLC) (CSC_WLC) (FISH_WLC)4 NMP22_WLC CSC_FISH_PDD NMP22_WLC CSC_FISH_PDD FISH_PDD FISH_PDD CSC_FISH_PDD FISH_PDD(NMP22_WLC) (CSC_WLC) (NMP22_WLC) (CSC_WLC) (FISH_WLC) (FISH_WLC) (FISH_WLC) (FISH_WLC)NMP22_WLCNMP22_WLC NMP22_WLC CSC_NMP22_PDDFISH_PDD CSC_NMP22_PDD5 FISH_PDD CSC_NMP22_PDD(FISH_WLC) (NMP22_WLC) (FISH_WLC) (NMP22_WLC) (NMP22_WLC) (NMP22_WLC) (NMP22_WLC) (NMP22_WLC)NMP22_PDDIMM_WLC IMM_WLC CSC_NMP22_PDDIMM_WLC CSC_NMP22_PDD6 IMM_WLC CSC_NMP22_PDD(IMM_WLC) (CSC_WLC) (IMM_WLC) (CSC_WLC) (IMM_WLC) (IMM_WLC) (CSC_WLC) (NMP22_WLC)7 CSC_WLC IMM_PDD CSC_WLC IMM_PDD CSC_WLC CSC_WLC IMM_PDD IMM_WLC(CSC_WLC) (IMM_WLC) (CSC_WLC) (IMM_WLC) (CSC_WLC) (CSC_WLC) (IMM_WLC) (IMM_WLC)8 CSC_CTL_WLC CSC_CTL_PDD CSC_CTL_WLC CSC_CTL_PDD CSC_CTL_WLC CSC_CTL_WLC CSC_CTL_PDD IMM_PDD(CSC_WLC) (CSC_WLC) (CSC_WLC) (CSC_WLC) (CSC_WLC) (CSC_WLC) (CTL_WLC) (IMM_WLC)9 CSC_CTL_WLC CSC_CTL_PDD CSC_CTL_WLC CSC_CTL_PDD CSC_CTL_WLC CSC_CTL_WLC CSC_CTL_PDD CSC_CTL_WLC(CTL_WLC) (CTL_WLC) (CTL_WLC) (CTL_WLC) (CTL_WLC) (CTL_WLC) (CSC_WLC) (CTL_WLC)10 NMP22_PDD FISH_PDD NMP22_PDD FISH_PDD NMP22_PDD NMP22_PDD FISH_PDD CSC_FISH_WLC(NMP22_WLC) (FISH_WLC) (NMP22_WLC) (FISH_WLC) (NMP22_WLC) (NMP22_WLC) (FISH_WLC) (FISH_WLC)Note: For true test results correct diagnosis and higher life-year values are better and for false test results incorrect diagnosis and lower cost values are better.

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4TABLE 41 Results of the deterministic model for the 20-year time horizon (per case)StrategynumberStrategyAveragecostIncrementalcostTruepositivecasesdetectedIncrementalnumberof casesdetected1 CTL_WLC (CTL_WLC) £1043 162 CTL_PDD (CTL_WLC) £1094 £51 20 4 £133 FISH_WLC (FISH_WLC) £1171 £77 27 7 £114 FISH_PDD (FISH_WLC) £1235 £64 35 8 £85 NMP22_WLC (NMP22_WLC) £1242 £6 24 –11 Dominated6 IMM_WLC (IMM_WLC) £1321 £86 30 –5 Dominated7 NMP22_PDD (NMP22_WLC) £1345 £109 32 –3 Dominated8 IMM_PDD (IMM_WLC) £1458 £223 39 4 £569 CSC_CTL_WLC (CTL_WLC) £1662 £204 30 –9 Dominated10 CSC_FISH_WLC (FISH_WLC) £1807 £349 33 –5 Dominated11 CSC_NMP22_WLC (NMP22_ £1851 £393 32 –7 DominatedWLC)12 CSC_CTL_PDD (CTL_WLC) £1859 £401 39 0 Dominated13 CSC_WLC (CSC_WLC) £1920 £462 25 –14 Dominated14 CSC_IMM_WLC (IMM_WLC) £1941 £483 34 –5 Dominated15 CSC_CTL_WLC (CSC_WLC) £1997 £539 30 –9 Dominated16 CSC_FISH_PDD (FISH_WLC) £2005 £547 43 4 £13717 CSC_FISH_WLC (CSC_WLC) £2042 £37 33 –10 Dominated18 CSC_NMP22_WLC (CSC_ £2070 £65 32 –11 DominatedWLC)19 CSC_PDD (NMP22_WLC) £2082 £77 33 –10 Dominated20 CSC_NMP22_PDD (NMP22_ £2089 £84 42 –1 DominatedWLC)21 CSC_IMM_WLC (CSC_WLC) £2105 £100 34 –9 Dominated22 CSC_CTL_PDD (CSC_WLC) £2145 £140 39 -4 Dominated23 CSC_IMM_PDD (IMM_WLC) £2195 £190 44 1 £19024 CSC_FISH_PDD (CSC_WLC) £2270 £75 43 –1 Dominated25 CSC_NMP22_PDD (CSC_ £2318 £123 42 –2 DominatedWLC)26 CSC_IMM_PDD (CSC_WLC) £2370 £175 44 0 DominatedResults without dominated and extendedly dominated options1 CTL_WLC (CTL_WLC) £1043 162 CTL_PDD (CTL_WLC) £1094 £51 20 4 £133 FISH_WLC (FISH_WLC) £1171 £77 27 7 £114 FISH_PDD (FISH_WLC) £1235 £64 35 8 £88 IMM_PDD (IMM_WLC) £1458 £223 39 4 £5616 CSC_FISH_PDD (FISH_WLC) £2005 £547 43 4 £13723 CSC_IMM_PDD (IMM_WLC) £2195 £190 44 1 £190Note: In this table the ICER is the incremental cost per additional true positive case detected.ICER95© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Assessment of cost-effectivenessProbability cost-effective1.00.90.80.70.60.50.40.30.2+ + + + + + + + ++ + + + + +0.1++ + + + + + + ++ + + ++ + + +0.0 + + + ++ + 0 10 20 30 40 50+++++ + +++++++++++++++Ceiling ratio (Rc) (£000)++++++++++CTL-WLC (CTL-WLC)CTL-PDD (CTL-WLC)FISH-PDD (FISH-WLC)IMM-PDD (IMM-WLC)CSC-FISH-WLC (FISH-WLC)CSC-PDD (CSC-WLC)CSC-IMM-PDD (IMM-WLC)CSC-IMM-PDD (CSC-WLC)FIGURE 35 Cost-effectiveness acceptability curves determined by society’s willingness to pay for a life-year for the eight strategies.greatly when either lower or higher prevalencerates were used in the analysis. However, Figure35 shows that CSC_FISH_PDD (FISH_WLC) hadan increased probability of being considered costeffectivewhen the prevalence rate increased to20%. For example, the probability of CSC_FISH_PDD (FISH_WLC) being considered the mostcost-effective strategy would be greater than 22%when society is willing to pay more than £20,000per extra life-year. The CEACs for these sensitivityanalyses are shown in Appendix 20.Changes in the sensitivity and specificityof flexible cystoscopyWhen the sensitivity and specificity of flexiblecystoscopy were increased, life-years associatedwith ‘flexible cystoscopy’ strategies increased andrelevant costs decreased. Results of the changes inthe sensitivity and specificity of flexible cystoscopyare presented in Table 44 and, as this table shows,the strategies involving flexible cystoscopygenerally become more likely to be considered costeffectiveas its diagnostic performance increases.Nonetheless, at perhaps the most plausibleincrease of 5% in sensitivity and specificity forflexible cystoscopy compared with those of WLCthe probabilities that strategies involving flexiblecystoscopy are cost-effective are not greatlychanged. The CEACs for these sensitivity analysesare shown in Appendix 21.Relative risk rate of progression ofbladder cancer comparing no treatmentwith treatmentIn the sensitivity analysis the speed of progressionand rate of mortality for those falsely diagnosedas negative and hence not treated were altered.As might be expected, reducing these rates woulddecrease the cost-effectiveness of those strategiesassociated with fewer false negatives. Hence, theprobability that CTL_WLC (CTL_WLC) would beconsidered cost-effective increased from 18% inthe base-case analysis (RR 2.56) to 28% when theRR was 1 and society’s willingness to pay for a lifeyearwas £20,000 (Table 45). The CEACs for thesesensitivity analyses are shown in Appendix 22.Relative risk rate of recurrence andprogression comparing PDD with WLCAs indicated in Chapter 4, PDD is more likely toreduce the recurrence and progression of bladdercancer, decreasing these rates, and would thereforeincrease the cost-effectiveness of strategiesassociated with it. FISH_PDD (FISH_WLC) hadan increased probability of being consideredcost-effective when the RRs of recurrence andprogression were decreased to 0.64 and 0.56respectively (Tables 46 and 47 respectively). TheCEACs for these sensitivity analyses are shown inAppendices 23 and 24 respectively.96

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4TABLE 42 Results of the deterministic model for the 20-year time horizon (per life-year)Strategynumber Strategy CostIncrementalcostLife-yearsIncrementalyears1 CTL_WLC (CTL_WLC) £1043 11.592 CTL_PDD (CTL_WLC) £1094 £51 11.60 0.01 £34233 FISH_WLC (FISH_WLC) £1171 £77 11.62 0.01 £5575 a4 FISH_PDD (FISH_WLC) £1235 £64 11.64 0.02 £27625 NMP22_WLC (NMP22_WLC) £1242 £6 11.61 –0.03 Dominated6 IMM_WLC (IMM_WLC) £1321 £86 11.62 –0.02 Dominated7 NMP22_PDD (NMP22_WLC) £1345 £109 11.63 –0.01 Dominated8 IMM_PDD (IMM_WLC) £1458 £223 11.65 0.01 £28,8649 CSC_CTL_WLC (CTL_WLC) £1662 £204 11.62 –0.03 Dominated10 CSC_FISH_WLC (FISH_WLC) £1807 £349 11.63 –0.02 Dominated11 CSC_NMP22_WLC (NMP22_ £1851 £393 11.62 –0.02 DominatedWLC)12 CSC_CTL_PDD (CTL_WLC) £1859 £401 11.65 0 Dominated13 CSC_WLC (CSC_WLC) £1920 £462 11.60 –0.04 Dominated14 CSC_IMM_WLC (IMM_WLC) £1941 £483 11.63 –0.02 Dominated15 CSC_CTL_WLC (CSC_WLC) £1997 £539 11.62 –0.03 Dominated16 CSC_FISH_PDD (FISH_WLC) £2005 £547 11.66 0.01 £60,28417 CSC_FISH_WLC (CSC_WLC) £2042 £37 11.63 –0.03 Dominated18 CSC_NMP22_WLC (CSC_ £2070 £65 11.62 –0.03 DominatedWLC)19 CSC_PDD (CSC_WLC) £2082 £77 11.63 –0.03 Dominated20 CSC_NMP22_PDD (NMP22_ £2089 £84 11.65 –0.01 DominatedWLC)21 CSC_IMM_WLC (CSC_WLC) £2105 £100 11.63 –0.03 Dominated22 CSC_CTL_PDD (CSC_WLC) £2145 £140 11.64 –0.01 Dominated23 CSC_IMM_PDD (IMM_WLC) £2195 £190 11.66 < 0.01 £309,256 a24 CSC_FISH_PDD (CSC_WLC) £2270 £75 11.66 0 Dominated25 CSC_NMP22_PDD (CSC_ £2318 £123 11.65 –0.01 DominatedWLC)26 CSC_IMM_PDD (CSC_WLC) £2370 £175 11.66 < 0.01 £237,863Results without dominated and extendedly dominated options1 CTL_WLC (CTL_WLC) £1043 11.592 CTL_PDD (CTL_WLC) £1094 £51 11.60 0.01 £34234 FISH_PDD (FISH_WLC) £1235 £141 11.64 0.04 £38068 IMM_PDD (IMM_WLC) £1458 £223 11.65 0.01 £28,86416 CSC_FISH_PDD (FISH_WLC) £2005 £547 11.66 0.01 £60,28426 CSC_IMM_PDD (CSC_WLC) £2370 £365 11.66 < 0.01 £270,375a Extendedly dominated.ICER97© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Assessment of cost-effectiveness98TABLE 43 Sensitivity analysis associated with prevalence rateProbability of cost-effectiveness for different thresholdvalues for society’s willingness to pay for a life-year (%)Deterministic resultsIncrementalcost/ life-year £10,000 £20,000 £30,000 £40,000 £50,000Incrementallife-yearsAveragelife-yearsIncrementalcostAveragecostStrategyBase case, prevalence = 5%CTL_WLC (CTL_WLC) £1043 11.59 21 18 17 16 15CTL_PDD (CTL_WLC) £1094 £51 11.60 0.01 £3423 11 10 9 9 9FISH_PDD (FISH_WLC) £1235 £141 11.64 0.04 £3806 20 17 16 17 17IMM_PDD (IMM_WLC) £1458 £223 11.65 0.01 £28,864 18 18 17 17 17CSC_FISH_PDD (FISH_WLC) £2005 £547 11.66 0.01 £60,284 17 18 18 18 18CSC_PDD (CSC_WLC) £2082 £77 11.63 –0.03 Dominated 2 4 5 5 6CSC_IMM_PDD (IMM_WLC) £2195 £190 11.66 < 0.01 Extendedly 9 14 15 15 16dominatedCSC_IMM_PDD (CSC_WLC) £2370 £365 11.66 < 0.01 £270,375 1 3 3 3 3Prevalence = 1%CTL_WLC (CTL_WLC) £319 11.79 35 30 29 28 27CTL_PDD (CTL_WLC) £349 £30 11.79 < 0.01 £12,120 5 4 4 4 4FISH_PDD (FISH_WLC) £499 £150 11.79 < 0.01 £51,884 18 15 14 14 13IMM_PDD (IMM_WLC) £680 £180 11.79 0 Dominated 13 13 13 13 12CSC_PDD (CSC_WLC) £1148 £648 11.78 –0.01 Dominated 16 19 19 21 21CSC_FISH_PDD (FISH_WLC) £1306 £806 11.79 –0.01 Dominated 3 4 5 6 6CSC_IMM_PDD (IMM_WLC) £1427 £928 11.78 –0.01 Dominated 10 14 15 16 16CSC_IMM_PDD (CSC_WLC) £1450 £951 11.78 –0.01 Dominated 0 0 0 0 0

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Probability of cost-effectiveness for different thresholdvalues for society’s willingness to pay for a life-year (%)Deterministic resultsIncrementalcost/ life-year £10,000 £20,000 £30,000 £40,000 £50,000Incrementallife-yearsAveragelife-yearsIncrementalcostAveragecostStrategyPrevalence = 10%CTL_WLC (CTL_WLC) £1951 11.34 15 12 11 11 11CTL_PDD (CTL_WLC) £2033 £82 11.37 0.03 Extendedly 11 10 10 9 9dominatedFISH_PDD (FISH_WLC) £2168 £217 11.45 0.11 £2018 19 16 16 15 15IMM_PDD (IMM_WLC) £2449 £281 11.47 0.02 £13,597 18 16 16 15 15CSC_FISH_PDD (FISH_WLC) £2899 £451 11.50 0.03 £17,555 22 23 22 22 22CSC_IMM_PDD (IMM_WLC) £3177 £278 11.50 0.01 £48,035 12 15 16 17 17CSC_PDD (CSC_WLC) £3271 £93 11.43 –0.08 Dominated 2 4 5 6 6CSC_IMM_PDD (CSC_WLC) £3545 £368 11.50 < 0.01 £235,672 2 4 5 5 5Prevalence = 20%CTL_WLC (CTL_WLC) £3765 10.85 7 5 4 4 4CTL_PDD (CTL_WLC) £3904 £139 10.90 0.06 Extendedly 9 8 7 7 7dominatedFISH_PDD (FISH_WLC) £4019 £254 11.07 0.22 £1150 19 17 16 16 15IMM_PDD (IMM_WLC) £4411 £392 11.12 0.05 Extendedly 19 19 18 18 18dominatedCSC_FISH_PDD (FISH_WLC) £4667 £648 11.17 0.11 £6148 26 25 24 23 22CSC_IMM_PDD (IMM_WLC) £5119 £452 11.19 0.02 £28,183 17 21 21 21 21CSC_PDD (CSC_WLC) £5628 £509 11.03 –0.16 Dominated 2 3 4 4 4CSC_IMM_PDD (CSC_WLC) £5871 £751 11.19 < 0.01 £233,858 1 4 6 8 999© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Assessment of cost-effectiveness100TABLE 44 Sensitivity analysis associated with changes to the sensitivity and specificity of flexible cystoscopyProbability of cost-effectiveness for different thresholdvalues for society’s willingness to pay for a life-year (%)Deterministic resultsIncrementalcost/ life-year £10,000 £20,000 £30,000 £40,000 £50,000Incrementallife-yearsAveragelife-yearsIncrementalcostAveragecostStrategyBase case, CSC = WLC (sens = 0.71, spec = 0.72)CTL_WLC (CTL_WLC) £1043 11.59 21 18 17 16 15CTL_PDD (CTL_WLC) £1094 £51 11.60 0.01 £3423 11 10 9 9 9FISH_PDD (FISH_WLC) £1235 £141 11.64 0.04 £3806 20 17 16 17 17IMM_PDD (IMM_WLC) £1458 £223 11.65 0.01 £28,864 18 18 17 17 17CSC_FISH_PDD (FISH_WLC) £2005 £547 11.66 0.01 £60,284 17 18 18 18 18CSC_PDD (CSC_WLC) £2082 £77 11.63 –0.03 Dominated 2 4 5 5 6CSC_IMM_PDD (IMM_WLC) £2195 £190 11.66 < 0.01 Extendedly9 14 15 15 16dominatedCSC_IMM_PDD (CSC_WLC) £2370 £365 11.66 < 0.01 £270,375 1 3 3 3 3CSC = WLC + 5.0% (sens = 0.76, spec = 0.77)CTL_WLC (CTL_WLC) £1043 11.59 22 19 18 17 16CTL_PDD (CTL_WLC) £1094 £51 11.60 0.01 £3423 12 11 10 10 10FISH_PDD (FISH_WLC) £1235 £141 11.64 0.04 £3806 20 18 17 17 17IMM_PDD (IMM_WLC) £1458 £223 11.65 0.01 £28,864 16 16 15 15 15CSC_FISH_PDD (FISH_WLC) £1952 £494 11.66 0.01 £55,236 15 16 16 17 17CSC_PDD (CSC_WLC) £1986 £34 11.63 –0.02 Dominated 3 5 6 6 6CSC_IMM_PDD (IMM_WLC) £2149 £197 11.66 0 Dominated 12 15 16 17 17CSC_IMM_PDD (CSC_WLC) £2293 £341 11.66 0 Dominated 1 1 2 2 2

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Probability of cost-effectiveness for different thresholdvalues for society’s willingness to pay for a life-year (%)Deterministic resultsIncrementalcost/ life-year £10,000 £20,000 £30,000 £40,000 £50,000Incrementallife-yearsAveragelife-yearsIncrementalcostAveragecostStrategyCSC = WLC + 10.0% (sens = 0.81, spec = 0.82)CTL_WLC (CTL_WLC) £1043 11.59 21 18 16 15 15CTL_PDD (CTL_WLC) £1094 £51 11.60 0.01 £3423 9 7 8 7 7FISH_PDD (FISH_WLC) £1235 £141 11.64 0.04 £3806 20 18 17 17 16IMM_PDD (IMM_WLC) £1458 £223 11.65 0.01 £28,864 17 16 17 17 16CSC_PDD (CSC_WLC) £1881 £423 11.64 0 Dominated 3 5 6 7 7CSC_FISH_PDD (FISH_WLC) £1892 £434 11.66 0.01 £49,145 17 19 19 20 20CSC_IMM_PDD (IMM_WLC) £2096 £204 11.66 0 Dominated 12 15 16 16 17CSC_IMM_PDD (CSC_WLC) £2211 £319 11.65 0 Dominated 1 1 2 2 2CSC = WLC + 25.0% (sens = 0.96, spec = 0.97)CTL_WLC (CTL_WLC) £1043 11.59 14 11 10 10 9CTL_PDD (CTL_WLC) £1094 £51 11.60 0.01 £3423 9 8 7 7 7FISH_PDD (FISH_WLC) £1235 £141 11.64 0.04 £3806 16 15 15 14 14IMM_PDD (IMM_WLC) £1458 £223 11.65 0.01 Extendedly17 17 18 18 18dominatedCSC_PDD (CSC_WLC) £1552 £317 11.67 0.03 £10,485 17 17 18 18 18CSC_FISH_PDD (FISH_WLC) £1706 £154 11.66 –0.01 Dominated 17 17 18 18 18CSC_IMM_PDD (IMM_WLC) £1931 £379 11.65 –0.02 Dominated 8 10 10 10 10CSC_IMM_PDD (CSC_WLC) £1964 £412 11.65 –0.02 Dominated 4 5 6 6 6Sens, sensitivity; spec, specificity.101© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Assessment of cost-effectiveness102TABLE 45 Sensitivity analysis associated with relative risk for progression comparing no treatment with treatment of bladder cancerProbability of cost-effectiveness for different thresholdvalues for society’s willingness to pay for a life-year (%)Deterministic resultIncrementalcost/life-year £10,000 £20,000 £30,000 £40,000 £50,000Incrementallife-yearsAveragelife-yearsIncrementalcostAveragecostStrategyBase case, RR = 2.56CTL_WLC (CTL_WLC) £1043 11.59 21 18 17 16 15CTL_PDD (CTL_WLC) £1094 £51 11.60 0.01 £3423 11 10 9 9 9FISH_PDD (FISH_WLC) £1235 £141 11.64 0.04 £3806 20 17 16 17 17IMM_PDD (IMM_WLC) £1458 £223 11.65 0.01 £28,864 18 18 17 17 17CSC_FISH_PDD (FISH_WLC) £2005 £547 11.66 0.01 £60,284 17 18 18 18 18CSC_PDD (CSC_WLC) £2082 £77 11.63 –0.03 Dominated 2 4 5 5 6CSC_IMM_PDD (IMM_WLC) £2195 £190 11.66 < 0.01 Extendedly 9 14 15 15 16dominatedCSC_IMM_PDD (CSC_WLC) £2370 £365 11.66 < 0.01 £270,375 1 3 3 3 3RR = 2.0CTL_WLC (CTL_WLC) £1040 11.63 24 22 21 20 19CTL_PDD (CTL_WLC) £1092 £52 11.64 0.01 £5707 11 9 9 9 8FISH_PDD (FISH_WLC) £1242 £150 11.66 0.02 £7115 20 18 17 17 17IMM_PDD (IMM_WLC) £1464 £222 11.66 0 £51,443 16 16 15 15 14CSC_FISH_PDD (FISH_WLC) £2008 £544 11.66 0 £127,281 16 18 18 19 19CSC_PDD (CSC_WLC) £2110 £102 11.64 –0.02 Dominated 3 4 5 6 6CSC_IMM_PDD (IMM_WLC) £2198 £190 11.66 0.00 Dominated 10 12 13 14 14CSC_IMM_PDD (CSC_WLC) £2378 £370 11.67 < 0.01 £897,929 1 2 2 3 3

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Probability of cost-effectiveness for different thresholdvalues for society’s willingness to pay for a life-year (%)Deterministic resultIncrementalcost/life-year £10,000 £20,000 £30,000 £40,000 £50,000Incrementallife-yearsAveragelife-yearsIncrementalcostAveragecostStrategyRR = 1.5CTL_WLC (CTL_WLC) £1026 11.66 30 26 24 23 23CTL_PDD (CTL_WLC) £1081 £55 11.66 0 £11,311 10 9 9 9 9FISH_PDD (FISH_WLC) £1243 £162 11.67 0.01 £24,861 19 18 16 16 16IMM_PDD (IMM_WLC) £1465 £223 11.67 < 0.01 £204,841 13 13 13 13 13CSC_FISH_PDD (FISH_WLC) £2008 £543 11.67 < 0.01 £2,836,313 15 16 17 17 18CSC_PDD (CSC_WLC) £2133 £124 11.66 –0.01 Dominated 3 5 6 6 6CSC_IMM_PDD (IMM_WLC) £2200 £191 11.67 0 Dominated 8 11 13 13 13CSC_IMM_PDD (CSC_WLC) £2384 £376 11.67 0 Dominated 2 2 2 3 3RR = 1.0CTL_WLC (CTL_WLC) £999 11.69 32 28 26 25 25CTL_PDD (CTL_WLC) £1066 £67 11.69 < 0.01 £123,479 11 10 10 10 10FISH_PDD (FISH_WLC) £1239 £173 11.68 –0.01 Dominated 17 15 14 14 14IMM_PDD (IMM_WLC) £1463 £398 11.68 –0.01 Dominated 14 12 13 13 12CSC_FISH_PDD (FISH_WLC) £2006 £940 11.68 –0.01 Dominated 14 17 17 17 17CSC_PDD (CSC_WLC) £2149 £1083 11.68 –0.02 Dominated 3 6 7 8 8CSC_IMM_PDD (IMM_WLC) £2200 £1134 11.68 –0.02 Dominated 8 11 12 12 12CSC_IMM_PDD (CSC_WLC) £2389 £1323 11.68 –0.02 Dominated 1 2 2 2 2103© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Assessment of cost-effectiveness104TABLE 46 Sensitivity analysis associated with relative risk for recurrence comparing PDD with WLCProbability of cost-effectiveness for different thresholdvalues for society’s willingness to pay for a life-year (%)Deterministic resultIncrementalcost/life-year £10,000 £20,000 £30,000 £40,000 £50,000Incrementallife-yearsAveragelife-yearsIncrementalcostAveragecostStrategyBase case (RR_R = 1.0)CTL_WLC (CTL_WLC) £1043 11.59 21 18 17 16 15CTL_PDD (CTL_WLC) £1094 £51 11.60 0.01 £3423 11 10 9 9 9FISH_PDD (FISH_WLC) £1235 £141 11.64 0.04 £3806 20 17 16 17 17IMM_PDD (IMM_WLC) £1458 £223 11.65 0.01 £28,864 18 18 17 17 17CSC_FISH_PDD (FISH_WLC) £2005 £547 11.66 0.01 £60,284 17 18 18 18 18CSC_PDD (CSC_WLC) £2082 £77 11.63 –0.03 Dominated 2 4 5 5 6CSC_IMM_PDD (IMM_WLC) £2195 £190 11.66 < 0.01 Extendedly 9 14 15 15 16dominatedCSC_IMM_PDD (CSC_WLC) £2370 £365 11.66 < 0.01 £270,375 1 3 3 3 3RR_R = 0.9CTL_WLC (CTL_WLC) £1043 11.59 21 18 16 16 16CTL_PDD (CTL_WLC) £1090 £47 11.60 0.01 £3301 11 10 9 9 9FISH_PDD (FISH_WLC) £1227 £136 11.64 0.04 £3650 20 18 18 18 17IMM_PDD (IMM_WLC) £1451 £224 11.65 0.01 £28,840 16 15 15 15 15CSC_FISH_PDD (FISH_WLC) £1999 £548 11.66 0.01 £60,008 18 20 20 21 21CSC_PDD (CSC_WLC) £2080 £81 11.63 –0.03 Dominated 3 4 5 5 6CSC_IMM_PDD (IMM_WLC) £2190 £191 11.66 < 0.01 Extendedly 11 14 15 15 15dominatedCSC_IMM_PDD (CSC_WLC) £2367 £368 11.66 < 0.01 £277,574 0 1 1 2 2

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Probability of cost-effectiveness for different thresholdvalues for society’s willingness to pay for a life-year (%)Deterministic resultIncrementalcost/life-year £10,000 £20,000 £30,000 £40,000 £50,000Incrementallife-yearsAveragelife-yearsIncrementalcostAveragecostStrategyRR_R = 0.8CTL_WLC (CTL_WLC) £1043 11.59 17 13 12 12 12CTL_PDD (CTL_WLC) £1084 £41 11.60 0.01 £2773 13 11 10 9 9FISH_PDD (FISH_WLC) £1215 £131 11.64 0.04 £3469 23 21 19 18 18IMM_PDD (IMM_WLC) £1439 £224 11.65 0.01 £28,553 15 16 16 15 15CSC_FISH_PDD (FISH_WLC) £1986 £547 11.66 0.01 £59,527 16 18 19 19 19CSC_PDD (CSC_WLC) £2071 £85 11.63 –0.03 Dominated 4 5 7 8 8CSC_IMM_PDD (IMM_WLC) £2177 £191 11.66 < 0.01 £267,929 11 14 15 16 17CSC_IMM_PDD (CSC_WLC) £2357 £180 11.66 < 0.01 £304,531 1 2 3 3 3RR_R = 0.64CTL_WLC (CTL_WLC) £1043 11.59 18 14 13 12 12CTL_PDD (CTL_WLC) £1075 £31 11.61 0.02 £2005 13 12 12 12 11FISH_PDD (FISH_WLC) £1197 £122 11.64 0.04 £3181 21 19 18 18 18IMM_PDD (IMM_WLC) £1420 £223 11.65 0.01 £28,083 18 18 17 17 17CSC_FISH_PDD (FISH_WLC) £1965 £545 11.66 0.01 £58,791 15 17 18 18 18CSC_PDD (CSC_WLC) £2058 £93 11.63 –0.03 Dominated 3 4 5 6 6CSC_IMM_PDD (IMM_WLC) £2156 £190 11.66 < 0.01 £239,084 11 15 16 16 17CSC_IMM_PDD (CSC_WLC) £2341 £185 11.66 < 0.01 £395,494 1 1 2 2 2105© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Assessment of cost-effectiveness106TABLE 47 Sensitivity analysis associated with relative risk for progression comparing PDD with WLCProbability of cost-effectiveness for different thresholdvalues for society’s willingness to pay for a life-year (%)Deterministic resultIncrementalcost/life-year £10,000 £20,000 £30,000 £40,000 £50,000Incrementallife-yearsAveragelife-yearsIncrementalcostAveragecostStrategyBase case (RR_P = 1.0)CTL_WLC (CTL_WLC) £1043 11.59 21 18 17 16 15CTL_PDD (CTL_WLC) £1094 £51 11.60 0.01 £3423 11 10 9 9 9FISH_PDD (FISH_WLC) £1235 £141 11.64 0.04 £3806 20 17 16 17 17IMM_PDD (IMM_WLC) £1458 £223 11.65 0.01 £28,864 18 18 17 17 17CSC_FISH_PDD (FISH_WLC) £2005 £547 11.66 0.01 £60,284 17 18 18 18 18CSC_PDD (CSC_WLC) £2082 £77 11.63 –0.03 Dominated 2 4 5 5 6CSC_IMM_PDD (IMM_WLC) £2195 £190 11.66 < 0.01 Extendedly 9 14 15 15 16dominatedCSC_IMM_PDD (CSC_WLC) £2370 £365 11.66 < 0.01 £270,375 1 3 3 3 3RR_P = 0.9CTL_WLC (CTL_WLC) £1043 11.59 20 16 15 14 14CTL_PDD (CTL_WLC) £1095 £51 11.60 0.01 £3697 12 10 9 9 9FISH_PDD (FISH_WLC) £1236 £141 11.64 0.04 £3797 21 19 19 19 19IMM_PDD (IMM_WLC) £1460 £225 11.65 0.01 £29,123 15 16 15 15 14CSC_FISH_PDD (FISH_WLC) £2009 £549 11.66 0.01 £60,273 15 17 18 18 19CSC_PDD (CSC_WLC) £2086 £77 11.63 –0.03 Dominated 4 6 7 7 8CSC_IMM_PDD (IMM_WLC) £2200 £191 11.66 < 0.01 Extendedly 11 14 15 15 16dominatedCSC_IMM_PDD (CSC_WLC) £2376 £366 11.66 < 0.01 £272,285 2 2 2 3 3

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Probability of cost-effectiveness for different thresholdvalues for society’s willingness to pay for a life-year (%)Deterministic resultIncrementalcost/life-year £10,000 £20,000 £30,000 £40,000 £50,000Incrementallife-yearsAveragelife-yearsIncrementalcostAveragecostStrategyRR_P = 0.8CTL_WLC (CTL_WLC) £1043 11.59 19 16 15 14 14CTL_PDD (CTL_WLC) £1093 £50 11.60 0.01 £3529 11 9 9 9 9FISH_PDD (FISH_WLC) £1233 £140 11.64 0.04 £3760 25 22 21 20 20IMM_PDD (IMM_WLC) £1458 £225 11.65 0.01 £29,122 16 16 15 15 15CSC_FISH_PDD (FISH_WLC) £2006 £548 11.66 0.01 £60,046 16 18 19 19 20CSC_PDD (CSC_WLC) £2085 £78 11.63 –0.03 Dominated 3 5 6 6 6CSC_IMM_PDD (IMM_WLC) £2197 £191 11.66 < 0.01 Extendedly 11 12 13 14 15dominatedCSC_IMM_PDD (CSC_WLC) £2374 £367 11.66 < 0.01 £273,525 1 2 2 2 2RR_P = 0.56CTL_WLC (CTL_WLC) £1043 11.59 22 18 17 17 17CTL_PDD (CTL_WLC) £1089 £46 11.60 0.01 £3148 10 9 9 9 8FISH_PDD (FISH_WLC) £1227 £137 11.64 0.04 £3671 21 19 18 17 17IMM_PDD (IMM_WLC) £1452 £225 11.65 0.01 £29,110 15 15 15 14 14CSC_FISH_PDD (FISH_WLC) £1999 £547 11.66 0.01 £59,522 16 18 18 19 19CSC_PDD (CSC_WLC) £2080 £81 11.63 –0.03 Dominated 4 6 6 7 7CSC_IMM_PDD (IMM_WLC) £2191 £192 11.66 < 0.01 Extendedly 10 13 15 15 15dominatedCSC_IMM_PDD (CSC_WLC) £2369 £370 11.66 < 0.01 £276,805 2 3 3 3 3107© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Assessment of cost-effectivenessDiscount rateAnother sensitivity analysis was conducted bychanging the discount rate. The cost-effectivenessof the different strategies did not markedly changewhen the discount rate was changed between 0%and 6% (Table 48). The CEACs for these sensitivityanalyses are shown in Appendix 25.Proportions in each prognostic risk groupfor non-muscle-invasive diseaseChanges to the proportions in each prognosticrisk group for non-muscle-invasive disease werealso considered (note that as the proportion in thelow-risk group was increased, the proportion in thehigh-risk group decreased). The likelihood thatCTL_WLC (CTL_WLC), IMM_PDD (IMM_WLC),FISH_PDD (FISH_WLC) or CSC_FISH_PDD(FISH_WLC) would be considered cost-effectivedid not change although some non-dominated ornon-extendedly dominated strategies in the basecaseanalysis became dominated or extendedlydominated (Table 49). The CEACs for thesesensitivity analyses are shown in Appendix 26.Starting age of population and timehorizonSensitivity analysis was used to investigate theeffects of changing the starting age of the patientpopulation or changing the number of years thatthe model was performed. None of these sensitivityanalyses altered the likelihood of a given strategybeing considered cost-effective (Table 50). However,as the time horizon was reduced, the incrementalcost per life-year gained for each non-dominatedstrategy increased. This is because the majorityof costs are incurred in earlier years but of courseas the time horizon increases it is possible to gainmore life-years. The CEACs for the sensitivityanalyses are shown in Appendix 27.Strategy used in follow-up and quality oflife measuresThe final sensitivity analyses performed involvedincluding the use of PDD in follow-up andconducting cost–utility analysis using the valuesreported in Table 38. The CEACs for these twosensitivity analyses are shown in Appendices 28and 29 respectively. These results did not changemuch and there was no strategy that was likely to beconsidered the most cost-effective as shown in Table51. It was noted that the strategies associated withflexible cystoscopy were dominated by others whenusing QoL measures.Subgroup analysesNo subgroup analyses were conducted because oflack of relevant data.Summary of resultsThe economic model presented in this chapterconsidered some strategies involving PDD, WLC,biomarkers, cytology and flexible cystoscopythat are potentially relevant for the diagnosisand follow-up of bladder cancer patients. Theeffectiveness data for diagnostic tests came fromthe effectiveness review. However, there were nodata available on the performance of flexiblecystoscopy alone or combined with cytologyor biomarkers. Therefore, the sensitivity andspecificity of flexible cystoscopy were assumedto be the same as those of WLC as it was likelythat flexible and rigid cystoscopies would identifysimilar types of cancer at the same rate. Plausiblechanges in this rate did not change the results toany extent. For the strategies relating to combinedtests it was assumed that flexible cystoscopy wascombined with cytology and/or biomarkers andthen followed by WLC or PDD if any one of theprevious tests performed was positive.The base-case analysis model suggests that, fora prevalence rate of 5% in a population withsuspected bladder cancer, the diagnostic strategythat would be cost-effective depends upon thevalue that society would be willing to pay to obtainan additional unit of outcome. Broadly speakingthe results based on cases detected were similar tothose based upon life-years. The strategy of flexiblecystoscopy and ImmunoCyt followed by PDD ininitial diagnosis and flexible cystoscopy followed byWLC in follow-up [CSC_IMM_PDD(CSC_WLC)],which produced 11.66 life-years and had a meancost of £2370 per patient, was the most costlyamong the diagnostic strategies in the base-caseanalysis. The CTL_WLC (CTL_WLC) strategy wasthe least costly (£1043) and least effective (11.59life-years). Although the differences betweenstrategies in terms of costs and effects appear tobe small, the important issue is the results of thewillingness to pay for additional gain. CTL_WLC(CTL_WLC) had a greater chance of being costeffectivewhen the willingness to pay was less than£20,000 per life-year. IMM_PDD (IMM_WLC),FISH_PDD (FISH_WLC) and CSC_FISH_PDD(FISH_WLC) had a greater probability of beingcost-effective when the willingness to pay was108

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4TABLE 48 Sensitivity analysis associated with discount rateProbability of cost-effectiveness for different thresholdvalues for society’s willingness to pay for a life-year (%)Deterministic resultIncrementalcost/life-year £10,000 £20,000 £30,000 £40,000 £50,000Incrementallife-yearsAveragelife-yearsIncrementalcostAveragecostStrategyBase case (discount rate = 3.5%)CTL_WLC (CTL_WLC) £1043 11.59 21 18 17 16 15CTL_PDD (CTL_WLC) £1094 £51 11.60 0.01 £3423 11 10 9 9 9FISH_PDD (FISH_WLC) £1235 £141 11.64 0.04 £3806 20 17 16 17 17IMM_PDD (IMM_WLC) £1458 £223 11.65 0.01 £28,864 18 18 17 17 17CSC_FISH_PDD (FISH_WLC) £2005 £547 11.66 0.01 £60,284 17 18 18 18 18CSC_PDD (CSC_WLC) £2082 £77 11.63 –0.03 Dominated 2 4 5 5 6CSC_IMM_PDD (IMM_WLC) £2195 £190 11.66 < 0.01 Extendedly 9 14 15 15 16dominatedCSC_IMM_PDD (CSC_WLC) £2370 £365 11.66 < 0.01 £270,375 1 3 3 3 3Discount rate = 6%CTL_WLC (CTL_WLC) £978 9.84 21 17 15 15 14CTL_PDD (CTL_WLC) £1031 £53 9.85 0.01 £4364 12 10 9 9 9FISH_PDD (FISH_WLC) £1166 £134 9.88 0.03 £4509 23 20 19 18 17IMM_PDD (IMM_WLC) £1382 £217 9.88 0.01 £36,596 16 14 14 14 14CSC_FISH_PDD (FISH_WLC) £1940 £558 9.89 0.01 £80,682 16 21 21 21 22CSC_PDD (CSC_WLC) £1987 £46 9.87 –0.03 Dominated 2 4 5 5 6CSC_IMM_PDD (IMM_WLC) £2122 £181 9.89 < 0.01 Extendedly 10 13 15 15 16dominatedCSC_IMM_PDD (CSC_WLC) £2278 £337 9.89 < 0.01 £408,489 1 2 2 3 3continued109© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Assessment of cost-effectiveness110TABLE 48 Sensitivity analysis associated with discount rate (continued)Probability of cost-effectiveness for different thresholdvalues for society’s willingness to pay for a life-year (%)Deterministic resultIncrementalcost/life-year £10,000 £20,000 £30,000 £40,000 £50,000Incrementallife-yearsAveragelife-yearsIncrementalcostAveragecostStrategyDiscount rate = 1%CTL_WLC (CTL_WLC) £1124 13.95 19 17 16 15 15CTL_PDD (CTL_WLC) £1172 £48 13.97 0.02 £2607 10 9 8 8 8FISH_PDD (FISH_WLC) £1323 £151 14.02 0.05 £3215 20 17 16 15 15IMM_PDD (IMM_WLC) £1554 £231 14.03 0.01 £22,648 17 16 16 16 15CSC_FISH_PDD (FISH_WLC) £2088 £533 14.04 0.01 £44,272 18 20 21 21 21CSC_PDD (CSC_WLC) £2203 £115 14.00 –0.04 Dominated 4 5 7 7 8CSC_IMM_PDD (IMM_WLC) £2289 £201 14.04 < 0.01 Extendedly 12 15 16 16 16dominatedCSC_IMM_PDD (CSC_WLC) £2487 £399 14.04 < 0.01 £190,983 1 2 2 3 3Discount rate = 0%CTL_WLC (CTL_WLC) £1162 15.13 17 15 14 14 14CTL_PDD (CTL_WLC) £1209 £47 15.15 0.02 £2316 9 8 8 8 8FISH_PDD (FISH_WLC) £1365 £156 15.20 0.05 £3009 21 19 17 17 17IMM_PDD (IMM_WLC) £1600 £235 15.21 0.01 £20,526 17 17 16 16 16CSC_FISH_PDD (FISH_WLC) £2127 £527 15.23 0.01 £38,899 17 18 18 18 18CSC_PDD (CSC_WLC) £2261 £134 15.18 –0.04 Dominated 4 6 7 7 7CSC_IMM_PDD (IMM_WLC) £2334 £207 15.23 < 0.01 £162,110 13 15 17 17 17CSC_IMM_PDD (CSC_WLC) £2543 £209 15.23 < 0.01 £174,776 2 3 3 3 4

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4TABLE 49 Sensitivity analysis associated with proportions in prognostic risk groupsProbability of cost-effectiveness for different thresholdvalues for society’s willingness to pay for a life-year (%)Deterministic resultsIncrementalcost/life-year £10,000 £20,000 £30,000 £40,000 £50,000Incrementallife-yearsAveragelife-yearsIncrementalcostAveragecostStrategyBase case, low = 0.1, high = 0.45CTL_WLC (CTL_WLC) £1043 11.59 21 18 17 16 15CTL_PDD (CTL_WLC) £1094 £51 11.60 0.01 £3423 11 10 9 9 9FISH_PDD (FISH_WLC) £1235 £141 11.64 0.04 £3806 20 17 16 17 17IMM_PDD (IMM_WLC) £1458 £223 11.65 0.01 £28,864 18 18 17 17 17CSC_FISH_PDD (FISH_WLC) £2005 £547 11.66 0.01 £60,284 17 18 18 18 18CSC_PDD (CSC_WLC) £2082 £77 11.63 –0.03 Dominated 2 4 5 5 6CSC_IMM_PDD (IMM_WLC) £2195 £190 11.66 < 0.01 Extendedly 9 14 15 15 16dominatedCSC_IMM_PDD (CSC_WLC) £2370 £365 11.66 < 0.01 £270,375 1 3 3 3 3Low = 0.3, high = 0.3CTL_WLC (CTL_WLC) £1020 11.60 21 17 15 15 15CTL_PDD (CTL_WLC) £1071 £51 11.61 0.01 Extendedly 9 9 8 8 8dominatedFISH_PDD (FISH_WLC) £1190 £170 11.65 0.05 £3254 22 20 19 19 18IMM_PDD (IMM_WLC) £1400 £210 11.66 0.01 £27,170 16 16 15 15 15CSC_FISH_PDD (FISH_WLC) £1957 £557 11.67 0.01 £58,259 18 19 20 20 21CSC_PDD (CSC_WLC) £2011 £54 11.64 –0.03 Dominated 4 6 7 7 7CSC_IMM_PDD (IMM_WLC) £2132 £175 11.67 < 0.01 £224,407 9 12 13 13 14CSC_IMM_PDD (CSC_WLC) £2283 £151 11.67 < 0.01 £389,886 2 2 3 3 3continued111© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Assessment of cost-effectiveness112TABLE 49 Sensitivity analysis associated with proportions in prognostic risk groups (continued)Probability of cost-effectiveness for different thresholdvalues for society’s willingness to pay for a life-year (%)Deterministic resultsIncrementalcost/life-year £10,000 £20,000 £30,000 £40,000 £50,000Incrementallife-yearsAveragelife-yearsIncrementalcostAveragecostStrategyLow = 0.6, high = 0.1CTL_WLC (CTL_WLC) £979 11.61 18 16 14 14 13CTL_PDD (CTL_WLC) £1029 £49 11.63 0.01 Extendedly 10 9 8 8 8dominatedFISH_PDD (FISH_WLC) £1111 £132 11.67 0.05 £2487 23 21 20 18 18IMM_PDD (IMM_WLC) £1302 £190 11.67 0.01 £28,973 17 17 16 16 16CSC_FISH_PDD (FISH_WLC) £1867 £565 11.68 0.01 £51,823 19 20 20 21 21CSC_PDD (CSC_WLC) £1883 £17 11.65 –0.03 Dominated 2 5 6 7 7CSC_IMM_PDD (IMM_WLC) £2036 £170 11.68 < 0.01 £296,812 8 11 12 12 12CSC_IMM_PDD (CSC_WLC) £2117 £80 11.68 < 0.01 £420,138 3 4 4 4 4

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4TABLE 50 Sensitivity analysis associated with starting age and time horizonProbability of cost-effectiveness for different thresholdvalues for society’s willingness to pay for a life-year (%)Deterministic resultsIncrementalcost/life-year £10,000 £20,000 £30,000 £40,000 £50,000Incrementallife-yearsAveragelife-yearsIncrementalcostAveragecostStrategyBase case, starting age 67 yearsCTL_WLC (CTL_WLC) £1043 11.59 21 18 17 16 15CTL_PDD (CTL_WLC) £1094 £51 11.60 0.01 £3423 11 10 9 9 9FISH_PDD (FISH_WLC) £1235 £141 11.64 0.04 £3806 20 17 16 17 17IMM_PDD (IMM_WLC) £1458 £223 11.65 0.01 £28,864 18 18 17 17 17CSC_FISH_PDD (FISH_WLC) £2005 £547 11.66 0.01 £60,284 17 18 18 18 18CSC_PDD (CSC_WLC) £2082 £77 11.63 –0.03 Dominated 2 4 5 5 6CSC_IMM_PDD (IMM_WLC) £2195 £190 11.66 < 0.01 Extendedly 9 14 15 15 16dominatedCSC_IMM_PDD (CSC_WLC) £2370 £365 11.66 < 0.01 £270,375 1 3 3 3 3Starting age 57 yearsCTL_WLC (CTL_WLC) £1095 13.34 21 18 16 16 15CTL_PDD (CTL_WLC) £1145 £50 13.35 0.02 £2821 10 9 9 8 8FISH_PDD (FISH_WLC) £1294 £149 13.40 0.04 £3376 25 22 21 20 20IMM_PDD (IMM_WLC) £1522 £228 13.41 0.01 £23,832 16 16 16 16 16CSC_FISH_PDD (FISH_WLC) £2061 £539 13.42 0.01 £47,517 17 19 21 21 21CSC_PDD (CSC_WLC) £2162 £101 13.38 –0.04 (Dominated) 2 3 4 5 5CSC_IMM_PDD (IMM_WLC) £2258 £197 13.42 < 0.01 Extendedly 8 11 12 13 14dominatedCSC_IMM_PDD (CSC_WLC) £2447 £386 13.42 < 0.01 £197,884 1 1 2 2 2continued113© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Assessment of cost-effectiveness114TABLE 50 Sensitivity analysis associated with starting age and time horizon (continued)Probability of cost-effectiveness for different thresholdvalues for society’s willingness to pay for a life-year (%)Deterministic resultsIncrementalcost/life-year £10,000 £20,000 £30,000 £40,000 £50,000Incrementallife-yearsAveragelife-yearsIncrementalcostAveragecostStrategyStarting age 77 yearsCTL_WLC (CTL_WLC) £951 8.84 23 19 18 18 18CTL_PDD (CTL_WLC) £1005 £54 8.85 0.01 Extendedly 11 10 9 9 9dominatedFISH_PDD (FISH_WLC) £1135 £184 8.87 0.04 £5041 20 17 16 16 15IMM_PDD (IMM_WLC) £1349 £214 8.88 0 £44,105 13 12 12 11 11CSC_FISH_PDD (FISH_WLC) £1911 £562 8.88 0.01 £100,340 17 18 18 19 19CSC_PDD (CSC_WLC) £1946 £35 8.86 –0.02 Dominated 3 6 6 7 8CSC_IMM_PDD (IMM_WLC) £2089 £178 8.88 < 0.01 Extendedly 12 16 17 18 18dominatedCSC_IMM_PDD (CSC_WLC) £2239 £328 8.89 < 0.01 £709,968 2 3 3 3 3Time horizon 10 yearsCTL_WLC (CTL_WLC) £983 8.37 25 19 16 15 14CTL_PDD (CTL_WLC) £1034 £51 8.38 0.01 Extendedly 11 11 10 10 10dominatedFISH_PDD (FISH_WLC) £1155 £121 8.40 0.02 £5255 26 21 21 20 19IMM_PDD (IMM_WLC) £1369 £213 8.40 < 0.01 £54,101 17 17 17 16 16CSC_FISH_PDD (FISH_WLC) £1926 £557 8.41 < 0.01 £121,083 13 18 19 20 21CSC_PDD (CSC_WLC) £1972 £46 8.39 –0.02 Dominated 2 4 6 6 7CSC_IMM_PDD (IMM_WLC) £2105 £179 8.41 0 Dominated 6 9 10 11 12CSC_IMM_PDD (CSC_WLC) £2259 £333 8.41 < 0.01 £4,183,060 0 1 1 1 2

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4TABLE 51 Sensitivity analysis associated with strategy and quality of life measuresProbability of cost-effectiveness for different thresholdvalues for society’s willingness to pay for a life-year (%)Deterministic resultsIncrementalcost/life-year £10,000 £20,000 £30,000 £40,000 £50,000Incrementallife-yearsAveragelife-yearsIncrementalcostAveragecostStrategyBase case, second test in follow-up is WLCCTL_WLC (CTL_WLC) £1043 11.59 21 18 17 16 15CTL_PDD (CTL_WLC) £1094 £51 11.60 0.01 £3423 11 10 9 9 9FISH_PDD (FISH_WLC) £1235 £141 11.64 0.04 £3806 20 17 16 17 17IMM_PDD (IMM_WLC) £1458 £223 11.65 0.01 £28,864 18 18 17 17 17CSC_FISH_PDD (FISH_WLC) £2005 £547 11.66 0.01 £60,284 17 18 18 18 18CSC_PDD (CSC_WLC) £2082 £77 11.63 –0.03 Dominated 2 4 5 5 6CSC_IMM_PDD (IMM_WLC) £2195 £190 11.66 < 0.01 Extendedly 9 14 15 15 16dominatedCSC_IMM_PDD (CSC_WLC) £2370 £365 11.66 < 0.01 £270,375 1 3 3 3 3Second test in follow-up is PDDCTL_WLC (CTL_PDD) £1069 11.59 20 17 15 15 14CTL_PDD (CTL_PDD) £1119 £50 11.60 0.01 £3372 11 9 9 8 8FISH_PDD (FISH_WLC) £1279 £160 11.64 0.04 £4314 20 18 17 17 17IMM_PDD (IMM_PDD) £1517 £238 11.65 0.01 £30,839 15 15 15 15 15CSC_FISH_PDD (FISH_PDD) £2051 £533 11.66 0.01 £58,765 18 20 21 21 21CSC_PDD (CSC_PDD) £2140 £90 11.63 –0.03 Dominated 4 5 6 6 6CSC_IMM_PDD (IMM_PDD) £2257 £207 11.66 < 0.01 Extendedly 10 13 15 15 15dominatedCSC_IMM_PDD (CSC_PDD) £2433 £383 11.66 < 0.01 £284,001 2 3 3 3 4continued115© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Assessment of cost-effectiveness116TABLE 51 Sensitivity analysis associated with strategy and quality of life measures (continued)Probability of cost-effectiveness for different thresholdvalues for society’s willingness to pay for a life-year (%)Deterministic resultsIncrementalcost/life-year £10,000 £20,000 £30,000 £40,000 £50,000Incrementallife-yearsAveragelife-yearsIncrementalcostAveragecostStrategyCost–utility analysisCTL_WLC (CTL_WLC) £1043 9.00 25 20 18 18 18CTL_PDD (CTL_WLC) £1094 £51 9.01 0.01 £4678 11 9 9 8 8FISH_PDD (FISH_WLC) £1235 £141 9.04 0.03 £5051 22 20 19 18 17IMM_PDD (IMM_WLC) £1458 £223 9.04 < 0.01 Extendedly 16 16 15 15 15dominatedCSC_FISH_PDD (FISH_WLC) £2005 £770 9.05 0.01 £66,905 16 19 19 19 19CSC_PDD (CSC_WLC) £2082 £77 9.01 –0.04 Dominated 2 3 4 4 4CSC_IMM_PDD (IMM_WLC) £2195 £190 9.05 0 Dominated 8 12 14 15 16CSC_IMM_PDD (CSC_WLC) £2370 £365 9.05 0 Dominated 1 2 2 3 3

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4increased to £30,000. Nevertheless, over mostof the range of willingness to pay values thereappeared to be no strategy that would have alikelihood of being cost-effective more than 50% ofthe time. For example, when the willingness to paywas over £10,000 per life-year the cost-effectivenessof FISH_PDD ranged from 16% to 20%. It shouldbe noted, however, that four out of the eightstrategies considered in the sensitivity analyseseach had a probability of being considered costeffectiveof approximately 20%. Three of these fourstrategies involved a biomarker and PDD.The results of probabilistic sensitivity analysesperformed to handle the uncertainty aroundthe parameters within the model were broadlyconsistent with the point estimates in the basecaseanalysis and did not change the order ofstrategies in terms of cost. The likelihood thatdifferent strategies might be considered costeffective,however, did change in some sensitivityanalyses. For example, the CSC_FISH_PDD(FISH_WLC) strategy had a 31% chance of beingconsidered cost-effective when the prevalence ratewas increased to 20% and society’s willingnessto pay for a life-year was £20,000. Furthermore,CSC_IMM_PDD (IMM_WLC) and CSC_FISH_PDD (FISH_WLC) had an increased chance ofbeing cost-effective in the situation in which thesensitivity and specificity of flexible cystoscopy wereincreased. This is important because in the basecaseanalysis it was assumed that the sensitivity andspecificity of flexible cystoscopy would be the sameas those of WLC. Both methods of cystoscopy usewhite light so it might be appropriate to assumethat they would identify (and miss) similar typesof cancer at the same rates. However, flexiblecystoscopy may be able to visualise more of thebladder than rigid cystoscopy. This means that itmay be possible for flexible cystoscopy to detectmore cancers. Whether this is true and, if it istrue, to what extent it improves sensitivity andspecificity is unclear. Overall, a potentially plausible5% gain in performance would not greatly alterthe conclusions drawn on the basis of the costeffectivenessresults.In sensitivity analyses the results did not changegreatly when the QoL estimates were used todetermine QALYs. The strategies associated withflexible cystoscopy were dominated and there wasa decreased chance of them being considered costeffective.This is because flexible cystoscopy, beingan invasive surgical procedure, is more likely toreduce QoL than cytology or biomarkers.In the model WLC was considered the secondtest in follow-up in each strategy if the result ofthe first test in follow-up was positive. Sensitivityanalysis suggested that the non-dominated ornon-extendedly dominated strategies had slightlyimproved life-years with higher costs comparedwith the base case when WLC in follow-up wasreplaced by PDD. However, strategies did notmarkedly change in how likely they were to be costeffective.117© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Chapter 7Assessment of factors relevant tothe NHS and other partiesFactors relevant to the NHSShould strategies that involve PDD be adoptedby the NHS then costs to the NHS would increaseand new capital equipment would be required.It is likely, however, that learning to use PDDshould be straightforward for an experiencedcystoscopist and hence the training period shouldbe short. Replacing WLC with PDD should increasethe number of cancers detected but this comesat the price of an increasing number of falsepositives. These false positives lead to an increasedworkload as unnecessary tests and investigationsare performed and, because these tests areunlikely to be without risk, a potential increase incomplications.The results of the economic evaluation suggest thatthe use of cytology as part of a diagnostic strategymight be reduced. Furthermore, the results suggestthat there may be merit in the increased use ofbiomarkers. Changes in the use of such tests wouldhave resource implications for the NHS and wouldsuggest transfers of resources between those partsof the NHS involved in the conduct, analysis andinterpretation of these tests.The adoption of less invasive tests in place of moreinvasive tests may also allow shifts in the balance ofcare between secondary and primary care, at leastfor initial diagnosis and potentially also for followup.Whether such changes are desirable wouldof course depend upon a host of other factors inaddition to feasibility, such as a desire to maintaincontinuity of care amongst those who have beentreated for bladder cancer.One consequence of any adoption of a moreeffective diagnostic test is that it may result ingreater survival (as estimated in the economicevaluation). Although this outcome is desirable itis important to remember that these patients willrequire continuing care and follow-up over a longerperiod. Therefore, it is possible that workload willincrease for those specialties involved in follow-up.Other longer-term effects, for example the effecton palliative services, are less easy to predict.© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.The results of the cost-effectiveness analysis suggestthat the strategies involving PDD were likely todetect more true positive cases and produce morelife-years at higher costs.Factors relevant to otherpartiesQuality of life for patientsThe use of strategies involving PDD, ImmunoCytand FISH could provide advantages to patientsin terms of early detection of disease and (forstrategies that replace an invasive procedure witha biomarker) provide a reduction in the number ofinvasive procedures that they may have to undergo.These strategies are also likely to decrease thenumber of false negatives, which will reduce therisks from false reassurance and the psychologicaldistress following a subsequent correct diagnosis.However, there is a price to pay for this in thatstrategies involving these tests are also associatedwith an increased chance of a false-positivediagnosis. Such a diagnosis may have health effectsas further tests and investigations performed arenot without risk. The false-positive diagnosis mayalso cause considerable anxiety and distress, notonly for the patients but also for their families.Patients and their families may also have viewsabout which diagnostic strategy they prefer thatgo beyond preferences over different aspects ofdiagnostic performance or longer-term healtheffects. In particular, there may be preferencesabout the process of care. All things being equalpatients would prefer the use of non-invasivebiomarker tests to the use of unpleasant, lessconvenient and potentially risky invasive tests.Nevertheless, all things are not equal and there arechoices and trade-offs to be made between process,short-term outcomes and long-term outcomes.Currently there are no data with which to informdecision-makers about how these differentoutcomes might be traded off against each other.119

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Chapter 8DiscussionStatement of principalfindingsPhotodynamic diagnosisDiagnostic accuracyThe included diagnostic accuracy studies reportedtrue and false positive and negative results orprovided information that allowed these data tobe calculated, thereby allowing further calculationof sensitivity, specificity, positive and negativelikelihood ratios, DORs and positive and negativepredictive values. Most studies compared PDDwith WLC. Studies comparing PDD with WLCwere included in the pooled estimates (metaanalyses)using a HSROC curve model. Thismethod takes into account the inherent tradeoffbetween sensitivity and specificity and alsoallows for differences in accuracy between studies.Summary pooled estimates of the sensitivity andthe specificity were calculated. Meta-analyses wereperformed on two levels:• patient• biopsy.In addition to the meta-analysis models of thediagnostic accuracy of PDD and WLC individually,two HSROC models were run for patient- andbiopsy-level analysis that simultaneously modelledPDD and WLC diagnostic accuracy from all of thestudies included in the pooled estimates. Analysiswas also undertaken on the sensitivity of PDDand WLC for the detection of stage and grade ofbladder cancer, which was considered in two broadcategories:• less aggressive, lower risk tumours (pTa, G1,G2)• more aggressive, higher risk tumours (pT1, G3,CIS).The sensitivity of PDD and WLC for the detectionof CIS alone was also considered. Stage andgrade analysis was undertaken for both patientandbiopsy-level detection of bladder cancer. Ananalysis of the sensitivity of PDD according to thetype of photosensitising agent used (5-ALA, HALor hypericin) was also undertaken. Information on© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.stage and grade analysis and type of agent used waspresented as median and range across studies.In terms of methodological quality, in all studiesthe spectrum of patients who received the tests wasconsidered to be representative of those who wouldreceive the tests in practice, partial verification biaswas avoided in that all patients who underwentPDD also received a reference standard test,and test review bias was avoided in that the PDDresults were considered to have been interpretedwithout knowledge of the results of the referencestandard test. However, in only 55% (15/27) ofstudies were patients considered to have receivedthe same reference standard regardless of the indextest result. All of the studies were judged to havesuffered from incorporation bias in that PDD wasnot considered to be independent of the referencestandard test as the biopsies used for the referencestandard were obtained via the PDD procedure.Although biopsy-level analysis of the accuracyof the test is more commonly reported, patientleveldata are more useful in determiningmanagement. Most studies took multiple biopsiesfrom participants, leading to clustering withinparticipants. We were unable to account for thisclustering in the biopsy-level analysis and thereforeestimates from the biopsy-level analysis will beto some degree artificially precise. In the pooledestimates for patient-level analysis, based on directevidence, PDD had higher sensitivity than WLC[92% (95% CI 80% to 100%) versus 71% (95% CI49% to 93%)] but lower specificity [57% (95% CI36% to 79%) versus 72% (95% CI 47% to 96%)]. Asfor patient-level analysis, in the pooled estimatesfor biopsy-level analysis, based on direct evidence,PDD also had higher sensitivity than WLC [93%(95% CI 90% to 96%) versus 65% (95% CI 55% to74%)] but lower specificity [60% (95% CI 49% to71%) versus 81% (95% CI 73% to 90%)]. In termsof sensitivity the upper CI for WLC did not overlapwith the lower CI for PDD, supporting evidenceof a difference in sensitivity in favour of PDD,and for specificity the upper CI for PDD did notoverlap with the lower CI for WLC, supportingevidence of a difference in specificity in favour ofWLC. The corresponding CIs for the patient-levelanalysis were wider because of the reduced number121

Discussion122of studies although the direction was consistent.Although at least four of the five studies includedfor patient-level analysis and at least nine of the14 studies included for biopsy-level analysis in thepooled estimates contained a mixture of patientswith a suspicion of bladder cancer and thosewith previously diagnosed non-muscle-invasivedisease, test performance in these groups wasnot reported separately. The formal comparisonof PDD and WLC in patient- and biopsy-basedanalysis supported strong evidence of a differencein sensitivity in favour of PDD and in specificity infavour of WLC.The consequence of underdiagnosis at a patientlevel would mean that a patient’s treatment pathmay be detrimentally affected (e.g. dischargedfrom follow-up or chanelled to an inappropriatelylow-risk follow-up pathway). The consequence ofunderdiagnosis at a biopsy level is that a patientmay have suboptimal treatment of their knownbladder cancer, for example by failure to removean occult lesion or failure to institute a therapybecause of underestimating the patient’s riskcategory (e.g. by failing to diagnose concomitantCIS).Across studies the median sensitivities (range)of PDD and WLC for detecting lower risk, lessaggressive tumours were broadly similar forpatient-level detection [92% (20% to 95%) versus95% (8% to 100%)], but sensitivity was higherfor PDD for biopsy-level detection [96% (88%to 100%) versus 88% (74% to 100%)]. However,for the detection of more aggressive, higherrisk tumours the median sensitivities of PDDfor both patient-level [89% (6% to 100%)] andbiopsy-level [99% (54% to 100%)] detection weremuch higher than those of WLC [56% (0% to100%) and 67% (0% to 100%) respectively]. Thesuperior sensitivity of PDD was also reflected inthe detection of CIS alone, both for patient-level[83% (41% to 100%) versus 32% (0% to 83%)] andbiopsy-level [86% (54% to 100%) versus 50% (0%to 68%)] detection. However, these results shouldbe interpreted with caution as, other than for PDDbiopsy-based detection of lower risk disease, therange of sensitivities for both tests was very wide.[It may also be useful to note that, although notmeeting the inclusion criteria for this review asinformation was not provided on false positives andtrue negatives, Schmidbauer and colleagues, 207 in aEuropean multicentre study (19 centres), reportedthat, of 83 patients with CIS lesions, CIS wasdetected in 80 (96%) by PDD (HAL) compared with64 (77%) by WLC.]In terms of the relative sensitivities of thephotosensitising agents used, for patient-leveldetection of bladder cancer, the median sensitivity(range) of 5-ALA was slightly higher than thatof HAL [96% (64% to 100%) versus 90% (53% to96%)] whereas HAL had higher specificity than5-ALA [81% (43% to 100%) versus 52% (33% to67%)]. This situation was also reflected in biopsybaseddetection, with 5-ALA associated with highersensitivity [95% (87% to 98%) versus 85% (76% to94%)] but lower specificity [57% (32 to 67%) versus80% (58 to 100%)] than HAL. One study, by Simand colleagues, 76 reporting biopsy-based detectionof bladder cancer, used hypericin, reportingsensitivity of 82% and specificity of 91%. Theseresults suggest that 5-ALA may be associated withslightly higher sensitivity than HAL and that HALhas higher specificity than 5-ALA, but this shouldbe interpreted with caution as a number of factorsother than the photosensitising agent used mayhave contributed to the sensitivity and specificityvalues reported by the studies.Twelve studies 51–53,61–63,65,71–73,78,81 involving1543 patients reported that there were no sideeffects or no serious side effects associatedwith the photosensitising agent used. Sevenstudies 50,57,66,67,71,76,77 involving 746 patients reported41 side effects associated with the agent (5-ALA,19; HAL, 21; hypericin, 1), none of which wasconsidered to be serious.No other systematic reviews of PDD for detectingbladder cancer or reporting effectiveness outcomessuch as tumour recurrence were identified.In summary, compared with WLC, PDD hashigher sensitivity (fewer false negatives) and sowill detect cases of bladder cancer that are missedby WLC. However, compared with WLC, PDD’slower specificity (more false positives) will result inadditional, unnecessary biopsies of non-canceroustissue being taken and sent for analysis. Reasonscited in the literature for PDD false-positive resultsinclude: (1) inexperience in using PDD, in whichthe application of tangential fluorescence lightmay cause fluorescence in normal urothelium, (2)simple hyperplasia, (3) lesions with inflammationor scarring after previous TURBT when PDDwas carried out within 6 weeks of the previousprocedure and (4) previous instillation therapywithin 3–6 months of PDD. 208,209 De Dominicis andcolleagues 53 noted that a greater number of falsepositivelesions were detected during the periodwhen the authors were still not sufficiently trainedin the PDD procedure, particularly in the first 15

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4patients. In terms of the detection of stage andgrade of tumour, the results suggest that PDD ismuch more sensitive than WLC in the detectionof more aggressive, higher risk tumours, and thesuperior performance of PDD is also reflected inthe detection of CIS alone. From a clinical pointof view, compared with WLC, the advantagesof PDD’s higher overall sensitivity in detectingbladder cancer and also its higher sensitivity indetecting more aggressive, higher risk tumourshave to be weighed against the disadvantages ofa higher false-positive rate leading to additional,unnecessary biopsies of normal tissue beingtaken and potentially additional unnecessaryinvestigations being carried out and the resultinganxiety caused to patients and their families.Recurrence/progression of diseaseJain and Kockelbergh 210 noted that the highrecurrence rate of superficial bladder cancer, up to70% at 5 years, was responsible for a huge workloadfor urologists and much inconvenience for patients.They stated that the recurrence rate at the firstcheck cystoscopy varied enormously, suggestingthat incomplete resection or failure to detect smalladditional tumours may be a risk factor. 210 Theevidence from the diagnostic accuracy part of thisreview suggests that PDD has a higher sensitivityfor the detection of bladder cancer than WLC.Therefore, compared with WLC, the use of PDDduring initial TURBT may be expected to result inlower recurrence and progression rates, given thatsome tumours, including more aggressive, higherrisk tumours such as CIS, that might be missed byWLC will be detected by PDD.For the assessment of PDD-assisted TURBTcompared with WLC in terms of effectivenessoutcomes such as recurrence and progression,this review focused on RCTs. Four RCTs (reportedin eight papers) involving 544 participants metthe inclusion criteria. In terms of methodologicalquality, in all four studies the groups wereconsidered to be similar at baseline in termsof prognostic factors, eligibility criteria for thestudies were specified and the length of followupwas considered adequate in relation to theoutcomes of interest. However, in all studies itwas unclear whether the sequence generation wasreally random or whether treatment allocation wasadequately concealed.When meta-analysis was undertaken, the resultswere reported using RR as the effect measure anda fixed-effect model in the absence of statisticalheterogeneity, otherwise a random-effects model© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.was used. Two studies 86,89 reported recurrence-freesurvival at 12 and 24 months. In pooled estimatesthe direction of effect for both time points favouredPDD, although the difference was statisticallysignificant only at the 24-month time point (RR1.37, 95% CI 1.18 to 1.59).Four studies 86,88,89,92 reported residual tumour rateat first cystoscopy following TURBT. In pooledestimates PDD was associated with both statisticallysignificantly fewer residual pTa tumours (RR 0.32,95% CI 0.15 to 0.70) and fewer residual pT1tumours (RR 0.26, 95% CI 0.12 to 0.57) than WLC(overall pooled estimate RR 0.37, 95% CI 0.20 to0.69). Two of the studies 86,88 also reported residualtumour according to grade (G1, G2 and G3).Pooled estimates for G1 (RR 0.13, 95% CI 0.03 to0.71) and G2 (RR 0.32, 95% CI 0.16 to 0.64) werestatistically significant in favour of PDD, with thedirection of effect for G3 favouring PDD withoutreaching statistical significance (RR 0.57, 95% CI0.21 to 1.56), and the overall pooled estimate wasstatistically significant in favour of PDD (RR 0.31,95% CI 0.18 to 0.53).Two studies 88,89 reported tumour recurrence rateduring follow-up (5 years and 8 years respectively).In pooled estimates the direction of effect favouredPDD without reaching statistical significance(RR 0.64, 95% CI 0.39 to 1.06). Both studies 88,89also reported tumour progression during theirrespective follow-up periods and again in thepooled estimates the direction of effect favouredPDD without reaching statistical significance (RR0.57, 95% CI 0.22 to 1.46).Two studies 86,88 reported time to recurrence, bothfavouring PDD. Babjuk and colleagues 86 reported amedian time to recurrence of 17.05 months for thePDD group and 8.05 months for the WLC group,whereas Daniltchenko and colleagues 88 reported amedian (range) time to recurrence of 12 (2 to 58)months for the PDD group and 5 (2 to 52) monthsfor the WLC group.In summary, the evidence from the RCTs 86,88,89,92suggests that, compared with WLC, the use of PDDduring TURBT results in a statistically significantand large reduction in residual pTa and pT1tumours, longer recurrence-free survival of patientsat 2 years following surgery and a longer intervalbetween TURBT and tumour recurrence. However,these results should be interpreted with caution asthey are based on data from only four small studies.Based on the limited evidence it is unclear whetherPDD compared with WLC is associated with lower123

Discussiontumour recurrence and progression rates in thelonger term. Also, as discussed in the sectionon uncertainties, the administration of adjuvantintravesical therapy varied across the studies,making it difficult to assess what the true addedvalue of PDD might be in reducing recurrence ratesin routine clinical practice.Biomarkers and cytologyThe included diagnostic accuracy studies reportedtrue and false positive and negative results orprovided information that allowed these data to becalculated, thereby allowing the further calculationof sensitivity and specificity, positive and negativelikelihood ratios, DORs and positive and negativepredictive values for the three included urinebiomarkers (FISH, ImmunoCyt and NMP22) andcytology. Meta-analyses were undertaken for eachof the individual biomarkers and cytology forpatient-based detection of bladder cancer usingthe HSROC model. Additional meta-analyseswere also undertaken on the subset of studiesincluded in the pooled estimates that directlycompared biomarkers with cytology. Analysis wasalso undertaken on the sensitivity of the biomarkersand cytology for the detection of stage and gradeof bladder cancer, which was considered in thetwo broad categories previously referred to (lessaggressive/lower risk tumours and more aggressive/higher risk tumours), and also for detection of CISalone.For each biomarker only those studies that wereconsidered to have a similar (‘common’) cut-off,which was generally taken to be the most frequentlyused cut-off across studies, were included in themeta-analyses. The common cut-off was also usedwhen studies reported results using a numberof different cut-offs. The following commoncut-offs were used: FISH, gains of two or morechromosomes or five or more cells with polysomyor four or more aneusomic of 25 counted cells;ImmunoCyt, at least one green or one redfluorescent cell; NMP22, 10 U/ml; urine cytology,cytologist subjective assessment.In terms of methodological quality, in all 71studies the reference standard (cystoscopy withhistological assessment of biopsied tissue) wasconsidered likely to correctly classify bladdercancer. In 99% (70/71) of studies the spectrum ofpatients receiving the tests was considered to berepresentative of those who would receive the testin practice, and incorporation bias was avoidedin that the reference standard was independentof the biomarker/cytology test. In 96% (68/71) ofstudies partial verification bias was avoided in thatall patients who received a biomarker/cytologytest also received a reference standard test, and in87% (62/71) of studies differential verification biaswas avoided in that all patients received the samereference standard regardless of the index testresult. However, only 69% (49/71) of studies wereconsidered to have given a clear definition of whatconstituted a positive result.Table 52 shows the pooled estimates (sensitivity,specificity, DORs) as well as the median (range)positive and negative predictive values acrossstudies for the biomarkers and cytology for patientbaseddetection of bladder cancer. In the pooledestimates, based on indirect evidence, sensitivitywas highest for ImmunoCyt at 84% (95% CI 77% to91%) and lowest for cytology at 44% (95% CI 38%to 51%). ImmunoCyt (84%, 95% CI 77% to 91%)had higher sensitivity than NMP22 (68%, 95%CI 62% to 74%), with the lack of overlap betweenthe CIs supporting evidence of a difference insensitivity in favour of ImmunoCyt over NMP22.FISH (76%, 95% CI 65% to 84%), ImmunoCyt(84%, 95% CI 77% to 91%) and NMP22 (68%,95% CI 62% to 74%) all had higher sensitivitythan cytology (44%, 95% CI 38% to 51%), andagain the lack of overlap of the CIs betweenthe three biomarkers and cytology supportedevidence of a difference in sensitivity in favour ofthe three biomarkers over cytology. This situationwas reversed for specificity, which was highest forcytology at 96% (95% CI 94% to 98%) and lowestfor ImmunoCyt at 75% (68% to 83%). Cytology(96%, 95% CI 94% to 98%) had higher specificitythan FISH (85%, 95% CI 78% to 92%), ImmunoCyt(75%, 95% CI 68% to 83%) or NMP22 (79%, 95%CI 74% to 84%), with the lack of overlap of theCIs between cytology and the three biomarkerssupporting evidence of a difference in specificity infavour of cytology over the biomarkers.DORs (95% CI) ranged from 8 (5 to 11) to 19 (6 to26), with higher DORs indicating a better abilityof the test to differentiate between those with andthose without bladder cancer. Based on the DORvalues, FISH and cytology performed similarlywell [18 (3 to 32) and 19 (11 to 27) respectively],ImmunoCyt slightly less so [16 (6 to 26)] andNMP22 relatively poorly [8 (5 to 11)]. However, asthe DOR confidence intervals for each of the testsall overlapped these results should be interpretedwith caution.124

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4TABLE 52 Summary of pooled estimate results and predictive values for biomarkers and cytology for patient-based detection of bladdercancerTestNumberof studiesNumberanalysedSensitivity(%) (95% CI)Specificity(%) (95% CI)DOR(95% CI)PPV (%),median(range)NPV (%),median(range)FISH 12 2535 76 (65 to 84) 85 (78 to 92) 18 (3 to 32) 78 (27 to 99) 88 (36 to 97)ImmunoCyt 8 2896 84 (77 to 91) 75 (68 to 83) 16 (6 to 26) 54 (26 to 70) 93 (86 to 100)NMP22 28 10,119 68 (62 to 74) 79 (74 to 84) 8 (5 to 11) 52 (13 to 94) 82 (44 to 100)Cytology 36 14,260 44 (38 to 51) 96 (94 to 98) 19 (11 to 27) 80 (27 to 100) 80 (38 to 100)Across studies the median (range) PPVs were 80%(27% to 100%) for cytology (36 studies), 78% (27%to 99%) for FISH (12 studies), 54% (26% to 70%)for ImmunoCyt (eight studies) and 52% (13% to94%) for NMP22 (28 studies). NPVs were 93% (86%to 100%) for ImmunoCyt, 88% (36% to 97%) forFISH, 82% (44% to 100%) for NMP22 and 80%(38% to 100%) for cytology. However, it should benoted that predictive values are affected by diseaseprevalence, which is rarely constant across studies.Five studies 80,126,127,131,150 reporting NMP22 usedthe BladderChek point of care test. Across thesestudies, using a cut-off of 10 U/ml for a positive testresult, the median (range) sensitivity and specificityfor patient-based detection of bladder cancerwere 65% (50% to 85%) and 81% (40% to 87%)respectively. This is broadly similar to the 68%(95% CI 62% to 74%) sensitivity and 79% (95% CI74% to 84%) specificity for the 28 studies includedin the pooled estimates.In terms of the detection of stage/grade oftumour, ImmunoCyt had the highest mediansensitivity across studies (81%) for the detectionof less aggressive/lower risk tumours whereasFISH had the highest median sensitivity acrossstudies (95%) for the detection of more aggressive/higher risk tumours. For detection of CIS themedian sensitivity across studies for both FISHand ImmunoCyt was 100%. Cytology had thelowest sensitivity across studies for detectingless aggressive/lower risk tumours (27%), moreaggressive/higher risk tumours (69%) and alsoCIS (78%). For each of the tests, the mediansensitivity across studies was consistently higherfor the detection of more aggressive/higher risktumours than for the detection of less aggressive,lower risk tumours. The results for the stage/grade analysis should be interpreted with caution,however, as they are based on a relatively smallnumber of studies for ImmunoCyt (n = 6) andFISH (n = 10), as are the results for the detection© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.of CIS (ImmunoCyt, n = 6; FISH, n = 8; NMP22,n = 11). Additionally, for all of the tests the rangeof sensitivities across the studies for detecting stage/grade (both lower and higher risk) and CIS wasvery wide.Some studies included in the pooled estimatesfor the individual tests also directly comparedtests, comparing FISH with cytology (five studies),ImmunoCyt with cytology (six studies) andNMP22 with cytology (16 studies). In each set ofcomparisons cytology had lower sensitivity buthigher specificity than the biomarker with whichit was being compared. ImmunoCyt had highersensitivity (82%, 95% CI 76% to 89%) than cytology(44%, 95% CI 35% to 54%), whereas cytology hadhigher specificity (94%, 95% CI 91% to 97%) thanImmunoCyt (85%, 95% CI 71% to 85%), with thelack of overlap of the CIs supporting evidence ofdifferences in sensitivity in favour of ImmunoCytand in specificity in favour of cytology. Similarly,NMP22 had higher sensitivity (70%, 95% CI 59%to 80%) than cytology (40%, 95% CI 31% to 49%),whereas cytology had higher specificity (97%, 95%CI 95% to 99%) than NMP22 (81%, 95% CI 74% to88%), with the lack of overlap of the CIs supportingevidence of differences in sensitivity in favour ofNMP22 and in specificity in favour of cytology. Thepooled estimates for the sensitivity and specificityof the tests in the direct comparison studies werebroadly similar to those reported for the individualtests. The formal comparison for a differencebetween tests supported a difference between bothImmunoCyt and NMP22, and cytology, but therewas no evidence for a difference between FISHand cytology. The latter finding was based upona small number of studies and therefore a realdifference may exist as implied by the results forthe individual tests, which were based upon a largernumber of studies.In studies reporting the sensitivity and specificity oftests used in combination, sensitivity was generally125

Discussion126higher but specificity lower for the combinedtests compared with the higher value of the twoindividual tests. Most combinations of tests werereported by only one or two studies apart from thecombination of ImmunoCyt and cytology, whichwas reported by eight studies.In studies specifically reporting unevaluable tests,rates were 6.1% (65/1059, five studies) for FISH,5% (279/5292, 10 studies) for ImmunoCyt and 2%(54/2566, six studies) for cytology. None of theNMP22 studies specifically reported unevaluabletests.A few other systematic reviews have reportedthe sensitivity and specificity of biomarkers andcytology for detecting bladder cancer (Table53). In a systematic review and meta-analysis ofbiomarkers for the surveillance monitoring ofpreviously diagnosed bladder cancer Lotan andRoehrborn 211 reported, amongst other biomarkers,ImmunoCyt, NMP22 and cytology. A systematicreview by Glas and colleagues 212 of tumour markersin the diagnosis of primary bladder cancerreported, amongst others, NMP22 and cytology.A systematic review by van Rhijn and colleagues 213of urine markers for bladder cancer surveillancereported, amongst others, FISH, ImmunoCyt,NMP22 and cytology. Our results for the sensitivityand specificity of FISH, ImmunoCyt, NMP22 andcytology were mostly similar to those reported bythe other reviews, other than we reported higherspecificity for FISH (85% compared with 70%),higher sensitivity for ImmunoCyt (84% comparedwith 67%) and slightly higher specificity forNMP22 (79% compared with 73%) than van Rhijnand colleagues, 213 respectively, and, for cytology,higher sensitivity than Lotan and Roehrborn 211 andvan Rhijn and colleagues 213 (44% compared with34% and 35% respectively) but lower sensitivitythan Glas and colleagues (44% compared with 55%respectively). 212Strengths and limitations ofthe assessmentDiagnostic accuracy/effectivenessIn terms of strengths, for PDD/WLC effectivenessoutcomes such as recurrence we focused only onRCTs. In biomarker/cytology case–control studiesin which the control group contained a proportionof completely healthy controls, the control groupwas reanalysed minus the healthy controls to try tomake it more representative of the types of peoplewho would receive the tests in practice. If this wasnot possible the study was excluded. Case–controlstudies in which the whole control group consistedof healthy volunteers were excluded.In terms of limitations, non-English languagestudies were excluded, as were biomarker studieswith fewer than 100 patients included in theanalysis. Cytology studies whose publication yearpredated the publication year of the earliestincluded biomarker study were excluded. Althoughmost studies contained a mixture of patients witha suspicion of bladder cancer and those with ahistory of previously diagnosed bladder cancer, fewstudies reported results for these groups separately.Only five of the 41 included NMP22 studies usedthe BladderChek point of care test.UncertaintiesDiagnostic accuracy/effectivenessPDD in the clinical pathwayPDD could potentially be used in conjunctionwith rigid WLC at different stages in the clinicalpathway, including initial diagnosis and treatmentand surveillance monitoring. As with rigid WLC,PDD is not only a diagnostic test but also involvestreatment in that during the procedure suspiciouslesions are not only identified but also removed.Although most of the studies included in thepooled estimates for both patient- and biopsylevelanalysis contained a mixture of patientswith a suspicion of bladder cancer and those withpreviously diagnosed non-muscle-invasive disease,test performance in these groups was not reportedseparately. In the pooled estimates for bothpatient- and biopsy-level analysis, PDD had highersensitivity than WLC but lower specificity. Acrossstudies the median sensitivities (range) of PDDand WLC for detecting lower risk, less aggressivetumours were broadly similar for patient-leveldetection but the sensitivity of PDD was higherthan that of WLC for biopsy-level detection.However, for the detection of more aggressive,higher risk tumours the median sensitivities ofPDD for both patient- and biopsy-level detectionwere much higher than those of WLC and thissuperior sensitivity of PDD was also reflected inthe detection of CIS alone. This suggests that theappropriate point in the clinical pathway for PDDto be used is in conjunction with rigid WLC duringthe initial TURBT, and possibly also in conjunctionwith rigid WLC during surveillance monitoring ofsome high-risk patients.

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4TABLE 53 Systematic reviews/meta-analyses reporting sensitivity and specificity of biomarkers included in the present reviewFISH ImmunoCyt NMP22 CytologySensitivity(%)No. ofstudiesSpecificity(%)Sensitivity(%)No. ofstudiesSpecificity(%)Specificity(%)Sensitivity(%)No. ofstudiesSpecificity(%)Sensitivity(%)No. ofstudies36 44(38 to 51)79(74 to 84)96(94 to 98)28 68(62 to 74)75(68 to 83)8 84(77 to 91)94(90 to 96)85(78 to 92)Present review 12 76(65 to 84)26 55(48 to 62)99(83 to 99)78(72 to 83)Glas 2003 212 – – – – – – 14 67(60 to 73)18 34(20 to 53)80(58 to 91)Lotan 2003 211 – – – 1 86 79 15 73(47 to 87)26 35(13 to 75)73(55 to 98)94(85 to 100)15 71(47 to 100)75(62 to 82)6 67(52 to 100)70(66 to 93)van Rhijn 4 792005 213 (70 to 86)Figures in parentheses are 95% CIs apart from the study by van Rhijn and colleagues, which reported medians and ranges.127© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

DiscussionIn the four studies reporting effectivenessoutcomes, PDD was used during the initialTURBT. Patients were randomised to WLC- orWLC- and PDD-assisted TURBT, 86,89,92 or WLC- orPDD-assisted TURBT. 88 In the studies by Babjukand colleagues, 86 Denzinger and colleagues 89and Kriegmair and colleagues 92 residual tumourin both groups was evaluated by WLC-assistedresection. However, in the study by Daniltchenkoand colleagues 88 residual tumour in both groupswas evaluated by PDD-assisted resection. In threestudies the patients were followed up using WLCand urinary cytology. 86,88,89 (As the aim of the studyby Kriegmair and colleagues 92 was to assess residualtumour 10–14 days following TURBT there was nolonger-term follow-up).Adjuvant chemotherapyAdjuvant single-dose chemotherapy administeredwithin the first 24 hours and ideally withinthe first 6 hours following TURBT is standardpractice in the UK and much of Europe and canreduce recurrence rates by up to 50% in the first2 years. However, the administration of adjuvantintravesical therapy varied across the four studiesreporting effectiveness outcomes. The study byKriegmair and colleagues 92 did not state whetherintravesical therapy was given. The study byDaniltchenko and colleagues 88 reported that noneof the patients received adjuvant intravesicaltherapy. In the study by Babjuk and colleagues 86none of the patients with grade 1 or grade 2tumours received intravesical therapy, whereas allthose with grade 3 tumours received intravesicalBCG immunotherapy. In the study by Denzingerand colleagues 89 patients with a solitary primarytumour staged pTaG1–G2 (low-risk group) did notreceive intravesical therapy, whereas those withmultifocal tumours staged pTaG1–G2 or pT1G1–G2 (intermediate-risk group) underwent mitomycintherapy and those with primary stage pT1G3,CIS or treatment failure with mitomycin (highriskgroup) received BCG therapy. In this study,although there were consistently fewer recurrencesfor PDD compared with WLC across all risk groups,the difference in recurrence rates between PDDand WLC was smaller in the intermediate- andhigh-risk groups, both of which received adjuvantintravesical therapy, than it was in the low-riskgroup. 89 The fact that adjuvant intravesical therapywas not given to all of the patients in all of thestudies makes it difficult to assess what the trueadded value of PDD might be in reducing bladdertumour recurrence rates in routine practice.Biomarker/cytology test performancein patients with a suspicion of bladdercancer and those with a history of nonmuscle-invasivediseaseIt is possible that the diagnostic accuracy of urinebiomarkers/cytology may differ in patients newlypresenting with a suspicion of bladder cancercompared with those with a previous history ofnon-muscle-invasive disease. Most of the includedstudies contained a mixture of patients witha suspicion of bladder cancer and those withpreviously diagnosed disease but did not reportresults for these groups separately. However,in a few of the studies included in the pooledestimates that reported patient-level analysis thewhole patient population consisted either of oneor other of these groups. Table 54 shows, for eachtest, the median (range) sensitivity and specificityacross studies containing those newly presentingwith symptoms of bladder cancer and those withpreviously diagnosed non-muscle-invasive disease.For each test, both sensitivity and specificity wereslightly higher for the studies containing patientsnewly presenting with symptoms of bladder cancer,although these results should be interpreted withcaution as they are based on limited evidence,especially for FISH and ImmunoCyt.Biomarkers as a replacement for cytologyIn the pooled estimates the lack of overlap of theCIs between the three biomarkers and cytologysupported evidence of the biomarkers’ superiorsensitivity over cytology. ImmunoCyt had thehighest sensitivity (84%, 95% CI 77% to 91%),followed by FISH (76%, 95% CI 65% to 84%) andNMP22 (68%, 95% CI 62% to 74%), with cytologyhaving the lowest sensitivity (44%, 95% CI 38% to51%). This situation was reversed for specificity,with the lack of overlap of the CIs between cytologyand the three biomarkers supporting evidenceof cytology’s superior specificity over all threebiomarkers. The specificity of cytology was 96%(95% CI 94% to 98%), compared with 85% (95% CI78% to 92%) for FISH, 79% (95% CI 74% to 84%)for NMP22 and 75% (95% CI 68% to 73%) forImmunoCyt. The question of whether biomarkersmight replace cytology depends on the relativeimportance of higher sensitivity (fewer falsenegativeresults) compared with higher specificity(fewer false-positive results). If the sensitivity ofthe test was seen as being more important than itsspecificity then a test such as ImmunoCyt couldbe regarded as a potential candidate for replacingcytology. However, if the specificity of the test128

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4TABLE 54 Biomarker/cytology test performance in patients with a suspicion of bladder cancer and those with previously diagnoseddiseaseTestSuspicion/previoushistory of BCNumberof studiesNumberanalysedSensitivity (%),median (range)FISH Suspicion of BC 1 497 69 78Previous history of BC 1 250 64 73ImmunoCyt Suspicion of BC 1 280 85 88Previous history of BC 1 326 81 75Specificity (%),median (range)NMP22 Suspicion of BC 4 1893 71 (56 to 100) 86 (80 to 87)Previous history of BC 7 4284 69 (50 to 85) 81 (46 to 93)Cytology Suspicion of BC 7 3331 44 (16 to 100) 99 (87 to 100)Previous history of BC 6 4195 38 (12 to 47) 94 (83 to 97)BC, bladder cancer.Values for sensitivity and specificity are medians and ranges across studies.was seen as being more important then cytologywould remain the test of choice, given its superiorspecificity over all three biomarkers. A highlysensitive test will have few false negatives, whereasa highly specific test will have few false positives. Inthe case of high-risk bladder cancer, for example,the consequences of a false-negative test result arepotentially great, whereas those of a false-positivetest result are relatively low, inasmuch as thesepatients are unlikely to progress to a significantlymorbid treatment without a further diagnostic test.Biomarkers as a replacement for flexiblecystoscopy in monitoring patients with ahistory of low-risk bladder cancerThere have been suggestions that, givenappropriate sensitivity, a biomarker might replacethe use of some flexible cystoscopy for monitoringpatients with a history of low-risk bladder cancer.In the pooled estimates the median (95% CI)sensitivity was 84% (77% to 91%) for ImmunoCyt,76% (65% to 84%) for FISH and 68% (62% to 74%)for NMP22. ImmunoCyt at 84% had the highestsensitivity but this may still be regarded as too lowfor its consideration as a replacement for flexiblecystoscopy. Messing and colleagues 111 stated thatfor all biomarkers the lowest sensitivity was fordetecting low-grade tumours, which would be ofconcern if these tests were used to replace somecystoscopic examinations for monitoring patientswith a history of low-risk bladder cancer. Also, astudy by Yossepowitch and colleagues 214 interviewed200 consecutive patients previously diagnosedwith non-muscle-invasive bladder cancer who wereundergoing outpatient flexible cystoscopy at followup.The authors reported that, of the 200 patients,75% would accept the results of a urine test as areplacement for cystoscopy only if it was capableof detecting more than 95% of recurrent bladdertumours. Anxiety associated with the possibility ofmissing cancer was given as the major determinantof the minimal accepted accuracy. 214 However,these findings may not take account of the fact thatcystoscopy itself may not have perfect sensitivity.Random biopsiesThere appears to be no general consensus onwhether random biopsies of normal-appearingareas of the bladder should be undertaken duringcystoscopy. Some authors 54,68 argue that flat lesionssuch as dysplasias and CIS may be difficult tovisualise and therefore random biopsies should beundertaken. Kiemeney and colleagues, 215 in a studyinvolving 854 patients with superficial bladdercancer, noted that random biopsies from normalappearingareas revealed important histologicalfindings that were of high prognostic value.However, Witjes and colleagues, 194 in a study of1026 patients, claimed that random biopsies wereof little value in determining patients’ prognosis.In a study by van der Meijden and colleagues, 216the authors stated that in approximately 90%of patients the biopsies of normal-appearingurothelium in patients with stage Ta or T1 bladdercancer showed no abnormalities and therefore didnot contribute to staging or to the correct choice ofadjuvant therapy following TURBT. Jichlinski andcolleagues 65 stated that random biopsies of normalurothelium remained a subject of controversyand did not recommend their use in the general129© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Discussion130population of patients with non-muscle-invasivebladder cancer.Cost-effectiveness analysisStatement of principal findingsThe base-case analysis was based on a 5%prevalence rate of bladder cancer regardlessof whether the cost-effectiveness measure waspresented in terms of either cost per true positivecase detected or cost per life-year. Flexiblecystoscopy and ImmunoCyt followed by PDD ininitial diagnosis and flexible cystoscopy followed byWLC in follow-up [CSC_IMM_PDD (CSC_WLC)],which produced on average 11.66 life-years andhad a mean cost of £2370 per patient, was the mostcostly among the diagnostic strategies consideredin this study. The CTL_WLC strategy was the leastcostly (£1043) and least effective (11.59 life-years).There were six ‘non-dominated’ or non-extendedlydominated strategies in the base-case model whenoutcomes were measured in terms of incrementalcost per life-year: CTL_WLC (CTL_WLC),CTL_PDD (CTL_WLC), FISH_PDD (FISH_WLC),IMM_PDD (IMM_WLC), CSC_FISH_PDD (FISH_WLC) and CSC_IMM_PDD (CSC_WLC). Althoughthe differences between these appear to be smallin terms of cost and effects, it is important toremember that in only 5% of patients in the basecaseanalysis would testing provide any gain. Theimportant issue is what society would be willing topay for additional gain. The base-case results ofthe economic model indicated that the diagnosticstrategy that would be cost-effective depends uponthe value that society would be willing to pay toobtain an additional life-year. Cytology followed byWLC as the initial diagnosis and follow-up usingthe same interventions [CTL_WLC (CTL_WLC)]had a greater chance of being cost-effective whenthe willingness to pay was less than £20,000 perlife-year. However, when the willingness to paywas increased to £30,000 per life-year IMM_PDD(IMM_WLC), FISH_PDD (FISH_WLC) andCSC_FISH_PDD (FISH_PDD) also had a greaterprobability of being cost-effective. Nevertheless,over most of the range of willingness to pay valuesthere appeared to be no strategy that would have alikelihood of being cost-effective more than 50% ofthe time. For example, when the willingness to paywas over £10,000 per life-year the cost-effectivenessof FISH_PDD (FISH_WLC) ranged from 16% to20%. Of note, however, is that four of the eightstrategies considered in the probabilistic analysiswere each associated with a 20% probability ofbeing considered cost-effective at a range of valuesthat society might be willing to pay. Three of thesefour strategies involved the use of a biomarker andPDD.Results of probabilistic sensitivity analysesperformed to handle the uncertainty aroundthe parameters within the model were broadlyconsistent with the point estimates in the basecaseanalysis and did not change the order ofstrategies in terms of cost. However, the likelihoodthat different strategies might be considered costeffectivechanged when some of the parameterswere varied. For example, the CSC_FISH_PDD(FISH_WLC) strategy had a 25% chance of beingconsidered cost-effective when the prevalence ratewas increased to 20% and society’s willingnessto pay for a life-year was £20,000. There wassome concern that, because of lack of data, theperformance of flexible cystoscopy might beunderestimated. Sensitivity analyses suggest thatplausible (but contentious) increases in diagnosticperformance would not alter the conclusionsdrawn.In the cost–consequence analysis presented as partof the economic evaluation it was shown that thedifferent strategies were likely to vary not only interms of long-term performance but also in termsof short-term diagnostic performance. It is likelythat patients will have preferences about thesedifferent short-term outcomes that would not bereflected in estimates of life-years or indeed inQALYs based upon standard generic instrumentssuch as the EQ-5D. Furthermore, as indicatedin Chapter 7 patients may also have preferencesabout the process of care (including the use ofnon-invasive tests). The net impact of includingthese other potential benefits is unclear at presentand might be considered as an area for furtherresearch.Strengths and limitationsThis work is important as it is the first study toevaluate the cost-effectiveness of the diagnosticand follow-up strategies in patients with bladdercancer. The analysis considered the use ofPDD, biomarkers and cytology, in a variety ofcombinations, using a decision tree and a Markovmodel.A structured literature search was performed toidentify existing economic analyses of the diagnosisand management of patients with bladdercancer. No studies were identified that directlycompared the interventions under consideration.The approach adopted in this study provides anexplicit, reproducible methodology with which to

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4consider the interventions under consideration.Based on the relevant guidelines and detaileddiscussion with clinical experts involved in thisstudy, the care pathways were developed to buildup the structure of an economic model. Themethods used to estimate the parameters used inthe model were explicit and systematic and soughtto identify the best available evidence.Although the methods adopted to obtain theparameter estimates sought to identify thebest evidence available, the results should beinterpreted with caution as there are uncertaintiesand assumptions made in the economic model. Forexample, there was no evidence of what happens topatients who have false-negative results. It is likelythat bladder cancers missed in initial diagnosiswould not be treated until later, resulting in therisk of faster progression of disease. It was alsodifficult to identify suitable data on how quicklyuntreated bladder cancer progresses comparedwith treated bladder cancer. In the model a RR ofprogression or mortality comparing no treatment(false-negative results) with treatment (treatment oftrue positives) was used. In the base-case analysisit was assumed that the rate of RRs for progression(to muscle-invasive disease) and mortality forpatients who did not receive treatment (i.e. thosefalsely diagnosed as negative) compared with thosewho did receive treatment (i.e. those correctlydiagnosed as positive) was 2.56. It should be noted,however, that the sensitivity analysis that addressedthis assumption had very little impact on the resultsbecause there were small differences in falsenegativecases (or proportions) between strategiesat the level of prevalence (5%) of bladder cancerconsidered in the base-case analysis, indicatingthat this variable might not be that important as adeterminant of cost-effectiveness.The model structure focused on the diagnosis andmanagement of bladder cancer. The costs andbenefits of identifying and treating other causes ofthe symptoms (e.g. upper urinary tract problems,etc.) that patients presented with have not beenincluded. The net effect of not including this in amodel is uncertain.Besides the uncertainties surrounding theparameter estimates there were several otherlimitations to the report. One of the limitations ofthe economic evaluation was that it was not possibleto perform analysis on the impact of diagnosisand treatment of bladder cancer on QoL as therewere no data based on a generic economic tool.Although QoL data for other urological cancers© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.were available, after discussion with clinical expertsthey were deemed not to be generalisable to thisgroup of patients. A simple sensitivity analysissuggested that the inclusion of QoL estimates maynot greatly change the results. However, furtherresearch to elicit relevant health rate utilities wouldbe useful.Another challenge was that it was not possibleto conduct subgroup analysis because of a lackof data relating to subgroups. The subgroupsconsidered in this study were number of tumourson first cystoscopic examination; type of tumour;tumour recurrence at the first 3-month cystoscopicexamination following TURBT; and diagnosticperformance of the different PDD photosensitisingagents. Also considered were types of tumour andtumour recurrence on diagnostic performance ofthe different categories of urine biomarker; andwhether the urine sample for urine biomarkers wasvoided or obtained by bladder wash. More data areneeded to perform these subgroup analyses.Another limitation was the lack of evidence onthe performance of flexible cystoscopy, althoughit is the most commonly used test in current UKpractice. The reasons for lack of evidence forflexible cystoscopy may be attributable to the factthat it is an invasive procedure purely based on thejudgement of the person performing it, makingit difficult to evaluate the subjective outcome.Sensitivity analysis showed that potentiallyplausible improvements in the performance offlexible cystoscopy may not be meaningful.Another limitation was the determination of themost appropriate value for the prevalence rateof bladder cancer in the population that presentswith various symptoms of bladder cancer. Thereis evidence that the prevalence rate may varydepending on the symptoms that the patientspresent with. Ideally the population in the modelshould have been based on patients who hadprimary bladder cancer without a cancer history.However, it was difficult to establish relevantnumbers from the review of effectiveness as theresults were based on both first-time presentationsas well as repeat patients. It can be argued thatthe prevalence rate considered in the model mayeither overestimate or underestimate the numberof people with primary bladder cancer. Sensitivityanalysis results indicated that the prevalence ratehas a big impact on the cost-effectiveness results. Ata low level of prevalence (e.g. 1%) it is most likelythat the least costly strategy [CTL_WLC (CTL_WLC)] would be cost-effective over most of the131

Discussionstrategy range for a cost per life-year that societymight be willing to pay. At higher prevalences (e.g.20%) it is more likely that the more costly but moreeffective strategies would be considered worthwhile.One implication of the sensitivity of the model toprevalence rates is that it suggests that should asubgroup of the population be identified that has ahigher expected prevalence rate then it is possiblethat more effective (but more costly) strategieswould be worthwhile for such patients. Furtherresearch could consider whether such subgroupscould be identified.The economic evaluation may suffer from otherlimitations in addition to those related to theevidence base. A number of assumptions weremade with respect to the way that the decision treeand the Markov model were constructed. Theseassumptions were mostly made because of the lackof data to populate the model. As mentioned inChapter 6, it was assumed that the cycle lengths forrisk groups were the same during follow-up. Giventhe different intensities of follow-up for differenttypes of bladder cancer, in practice there would bemore than one opportunity per cycle for recurrentcancer to be diagnosed for some risk groups.A further assumption was made regarding themanagement of patients following recurrentdisease. During follow-up following treatment forbladder cancer, individuals could be incorrectlyidentified as still clear of cancer at a follow-upvisit (i.e. be a false-negative). There were no datato help model the impact of missing a cancer onfollow-up on mortality and progression. However,in our model all patients would have relativelyfrequent repeat testing during follow-up so theimpact of this limitation is debatable.Uncertainty in cost-effectivenessAlthough cost-effectiveness analysis was performedusing the best available data there was someuncertainty surrounding some of the parametersused in the model. One of these parameters wasthe risk group categorisation of non-muscleinvasivedisease. The ideal categorisation wouldneed to be based on all six prognostic risk factorsand include long-term survival and disease-freeinformation. As mentioned in Chapter 1, althoughthe EORTC classification was the most recentlyrecommended version and may have been the idealone to be adopted in the model, it was not possibleto use because of its complexity. Also, there wereno reliable data associated with the risk groups. Inaddition, the diagnostic technology for follow-upof bladder cancer may depend on the risk level forprogression and recurrence, for example T1G3and CIS will always be followed up using rigidcystoscopy. It is acknowledged that the definitionof risk groups may affect the judgement of costeffectivenessin the model. However, the sensitivityanalysis suggested that there is only a slight impacton base-case analysis when the proportions of riskgroup are changed.There was also uncertainty relating to survival andrecurrence-free and progression-free survival dataas they were only available up to 5 years post initialdiagnosis. These data were extrapolated to predictcost-effectiveness up to 20 years. Data at 5 yearssuggested little difference in terms of survival andrecurrence- and progression-free survival. However,results would be greatly strengthened if longertermrandomised data were available. For thepurposes of the model the mortality, progressionand recurrence rates were assumed to be constantover time. Given that data were extrapolatedfor 20 years in total, this assumption is perhapsunrealistic. However, it is unlikely that the effect ofholding the recurrence, progression and mortalityrates constant would have any impact on thedirection of results.The cost data used were also imprecise becausethe costs of diagnosis and treatments were mainlyidentified from NHS reference costs. As mentionedthere were very few studies that collected data onresource utilisation and, what published data therewere, were not generalisable to the UK. A furtherissue regarding costs was that inflation was nottaken into account. For the purposes of the analysisall prices were taken for the year 2007. However,the costs identified from NHS reference costs,the paper by Rodgers and colleagues 179 and theunpublished report for PDD were all 2006 costs.Normal practice within an economic evaluationwould argue that such costs be inflated to the samebase year allowing all costs to be comparable. Theanalyses conducted as part of this review, however,did not take into account inflation over time.However, it is anticipated that the failure to inflatethe costs, given the similar price years of the data,may have little impact on the results.One final point of uncertainty was the discountrate. The discount rates utilised followed publishedguidance relevant at the time that the technologyassessment report was commissioned. Increasesto the discount rate (mentioned in the methods132

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4chapter) would not change the overall directionof effects but are likely to make the more effectivestrategies (in terms of life-years) less likely to becost-effective. This is because these additionalbenefits accrue over time and hence are given lessweight when the discount rate is increased.133© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Chapter 9ConclusionsImplications for serviceprovisionIn terms of test performance, PDD has highersensitivity than WLC [pooled estimates forbiopsy-level analysis: 93% (95% CI 90% to 96%)versus 65% (95% CI 55% to 74%) respectively] indetecting bladder cancer in patients with symptomssuch as haematuria and is better at detectingmore aggressive, higher risk tumours, includingCIS [median (range) sensitivity across studiesfor biopsy-level analysis: 99% (54% to 100%)versus 67% (0% to 100%) respectively]. However,PDD has lower specificity than WLC [pooledestimates for biopsy-level analysis: 60% (95% CI49% to 71%) versus 81% (95% CI 73% to 90%)respectively]. The advantages of higher sensitivity(fewer false-negative results, better detection ofhigher risk tumours) have to be weighed againstthe disadvantages of lower specificity (more falsepositiveresults, leading to additional unnecessarybiopsies and potentially additional unnecessaryinvestigations and the resulting anxiety caused topatients and their families).In terms of the photosensitising agents used, acrossstudies the median (range) specificity reportedfor HAL was higher than that of 5-ALA for bothpatient-level [81% (43% to 100%) compared with52% (33 to 67%)] and biopsy-level [80% (58% to100%) compared with 57% (32% to 67%)] detectionof bladder cancer, although the ranges were wideand factors other than the agent used may alsohave contributed to the specificity values reported.Compared with WLC, the use of PDD at TURBTresults in fewer residual tumours at checkcystoscopy (pooled estimate RR 0.37, 95% CI0.20 to 0.69) and longer recurrence-free survival(pooled estimate RR 1.37, 95% CI 1.18 to 1.59),although these results are based on limitedevidence (three and two studies respectively) andshould be interpreted with caution. The advantagesof PDD at TURBT in reducing tumour recurrence(pooled estimate RR 0.64, 95% CI 0.39 to 1.06)and progression (pooled estimate RR 0.57, 95%CI 0.22 to 1.46) in the longer term were less clear(based on two studies, one with 5 years’ and onewith 8 years’ follow-up). In addition, as adjuvant© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.single-dose intravesical therapy following TURBT(standard practice in the UK and much of Europe)was not given to all of the patients in all of thestudies it is difficult to assess what the true addedvalue of PDD over WLC might be in routine clinicalpractice in terms of outcomes such as residualtumour at check cystoscopy, tumour recurrenceand progression. However, single-dose intravesicalchemotherapy is known to be ineffective againsthigh-risk tumours, the types more likely to bedetected by PDD.All three biomarkers had higher sensitivity butlower specificity than cytology for detectingbladder cancer in patients with symptoms suchas haematuria. In the pooled estimates (95% CI)ImmunoCyt had the highest sensitivity [84% (77%to 91%)], followed by FISH [76% (65% to 84%)],NMP22 [68% (62% to 74%)] and cytology [44%(38% to 51%)], whereas cytology had the highestspecificity [96% (94% to 98%)], followed by FISH[85% (78% to 92%)], NMP22 [79% (74% to 84%)]and ImmunoCyt [75% (68% to 83%)]. ImmunoCyt[84% (95% CI 77% to 91%)] had higher sensitivitythan NMP22 [68% (95% CI 62% to 74%)], withthe lack of overlap between the CIs supportingevidence of a difference in sensitivity in favour ofImmunoCyt. FISH [76% (95% CI 65% to 84%)] alsohad higher sensitivity than NMP22 although thedifference in sensitivity was more uncertain as theCIs overlapped. All three biomarkers and cytologywere better at detecting more aggressive, higherrisk tumours [median (range) sensitivity acrossstudies: FISH 95% (50% to 100%), ImmunoCyt90% (67% to 100%), NMP22 83% (0% to 100%),cytology 69% (0% to 100%)] than lower risk,less aggressive tumours [ImmunoCyt 81% (55%to 90%), FISH 65% (32% to 100%), NMP2250% (0% to 86%), cytology 27% (0% to 93%)]. Aurine biomarker test such as ImmunoCyt couldpotentially replace some cytology tests if highersensitivity (fewer false negatives) was consideredmore important than higher specificity (fewerfalse positives). However, if higher specificity wasconsidered to be more important then cytologywould remain the test of choice.The most cost-effective strategy for diagnosis andfollow-up of bladder cancer patients amongst PDD,135

Conclusions136WLC, biomarkers, cytology and flexible cystoscopywas evaluated. Based on currently available dataand taking into account the assumptions made inthe model, the strategy of flexible cystoscopy andImmunoCyt followed by PDD in initial diagnosisand flexible cystoscopy followed by WLC in followupis likely to be the most costly and the mosteffective (£2370 per patient and 11.66 life-years).The strategy of cytology followed by WLC in initialdiagnosis and follow-up is likely to be the leastcostly (£1043 per patient) and least effective interms of life-years (11.59) per patient. Comparedwith WLC in each strategy, PDD is more likelyto be cost-effective. However, it should be notedthat the diagnostic strategy that would be costeffectivedepends upon the value that societywould be willing to pay to obtain an additionallife-year. There appeared to be no strategy thatwould have a likelihood of being cost-effectivemore than 50% of the time over most of the rangeof willingness to pay values. Nevertheless, the fourstrategies involving PDD and biomarkers werecumulatively associated with over a 70% likelihoodof being considered cost-effective. The strategiesof ImmunoCyt or FISH followed by PDD in initialdiagnosis and ImmunoCyt or FISH followed byWLC in follow-up may be considered to be themost cost-effective when the willingness to pay isover £20,000.In summary, given the evidence presented ajudgement needs to be made as to whetherthe current ‘standard’ strategies with regard todiagnosis and follow-up of bladder cancer shouldbe altered. Currently, there is no standard strategyfor the detection and follow-up of primary bladdercancer. The implications of the finding thatdiagnostic strategies involving ImmunoCyt orFISH and PDD appear to have potential long-termoutcome benefits compared with current commonlyused strategies involving cytology or flexiblecystoscopy need to be considered. Diagnosticstrategies involving ImmunoCyt or FISH and PDDmay also have potential short-term benefits, suchas more true-positive cases detected and less falsenegativecases missed. However, any decision needsto take into account the extra costs associated withPDD and indeed whether the probable gains inQoL justify this increased cost.In the sensitivity analyses no strategy was likely tohave more than a 50% probability of being costeffective.This suggests that either the evidencebase is insufficient to warrant a change in practiceor we are indifferent between several strategiesin terms of cost-effectiveness, or more likely acombination of these two factors.There were no data on the combination of flexiblecystoscopy and cytology, the tests that are involvedin current commonly used strategies. Also, as therewere no data available with which to explicitlyincorporate QoL within the model, a judgementneeds to be made as to whether the expected gainin QoL is sufficient to offset any extra cost.Currently, PDD is used in only a few centres inthe UK and therefore the impact on the use ofoperating theatres arising from an increase in theuse of PDD would need to be considered. Learningto use PDD should be straightforward for anexperienced cystoscopist and the training periodshould be relatively short.Suggested researchprioritiesFurther research is required in the following areas:• RCTs comparing PDD with rigid WLC plusadjuvant intravesical therapy at TURBTin patients presumed to have non-muscleinvasivebladder cancer. The design of suchstudies should take into account participantcharacteristic risk groups, for example smokingand age, and allow outcomes to be reportedbased on risk categories at randomisation.Clinical effectiveness outcomes should includeresidual tumour rates at first check cystoscopy,recurrence-free survival, tumour recurrencerates, time to first recurrence, and progression.Such studies should make provision for longertermfollow-up (up to 10 years) and as a matterof course include an economic evaluationand measurement of health state utilities forincorporation into a cost–utility analysis.• Diagnostic cross-sectional studies comparingFISH with ImmunoCyt, NMP22 BladderChekpoint of care test and voided urine cytology,and also combinations of these tests, againsta reference standard of cystoscopy withhistological assessment of biopsied tissue inthe same patient population. The patientpopulation would be those newly presentingwith symptoms suspicious for bladder cancerand those with previously diagnosed nonmuscle-invasivebladder cancer. The studiesshould report true and false positives andnegatives for a patient-level analysis of thewhole patient group and also for the suspicionof bladder cancer/previously diagnoseddisease subgroups. For each of these groupsthe studies should report the sensitivity of thetests in detecting stage (pTa, pT1, ≥ pT2, CIS)

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4and grade (G1, G2, G3) of tumour, and size(< 1 cm, 1–3 cm, > 3cm) and number (one, twoto three, more than three) of tumours. Uppertract end points should also be considered.Observer variability in the interpretation oftests should also be reported. There shouldbe formal follow-up of patients who arecategorised as negative for bladder cancer tobetter understand the consequences of falsenegativecase ascertainment. The results ofsuch studies should be incorporated into arefined economic model that fully reflects thepragmatic factors listed above.• In addition, BAUS and the Renal Associationhave recently produced a new diagnosticalgorithm for the diagnosis of patients withhaematuria. This would be an appropriatesetting for further evaluating novel urinarybiomarkers such as ImmunoCyt and FISHand also for assessing their performance inspecific populations with a higher prevalenceof bladder cancer, such as men aged over 60years who smoke.• The level of QoL data suitable forincorporation into an economic model.Consideration should be given to the collectionof data suitable to expand on economicevaluations from cost-effectiveness analyses.Such data may be derived from furtherprospective studies or stand-alone studies thatseek to identify health state utilities relevant toa refined economic model.• The different strategies differ in terms oflonger-term outcomes and also in terms of theprocess of care and short-term outcomes. Thissuggests that consideration should be given topreference elicitation studies using recognisedmethodology that explore the trade-offs andvaluations between processes and healthoutcomes. Such analysis should be conductedin such a way that it can be incorporated intofuture models based on trial-based analysis.• False-negative results, either at diagnosis or atfollow-up, will prevent or at least delay thosepatients from receiving potentially beneficialtreatment. Further information is requiredas to what would happen to these patientsin practice and the impact of an incorrectdiagnosis on future survival, QoL and costs.Such information could be identified throughfollow-up of patients who are dischargedfollowing an initial negative result.137© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4AcknowledgementsWe thank Satchi Swami for advice on clinicalaspects of the review, Adrian Grant forcommenting on drafts, Clare Robertson andSusan Wong for assistance with assessing full-textstudies for inclusion, data extraction and qualityassessment, Clare Robertson for preparation ofthe characteristics of the included studies tablesand Kathleen McIntosh and Karen McLeodfor secretarial support. We thank the BritishAssociation of Urological Surgeons (BAUS)Section of Oncology Executive Committee forits suggestions as to which biomarkers the reviewmight consider. The Health Services Research Unitand Health Economics Research Unit, Institute ofApplied Health Sciences, University of Aberdeen,are core funded by the Chief Scientist Office ofthe Scottish Government Health Directorates. Theviews expressed are those of the authors and notnecessarily those of the funding bodies. Any errorsare the responsibility of the authors.Contribution of authorsinclusion, undertook data extraction and qualityassessment, drafted the chapters on photodynamicdiagnosis and biomarkers, and coordinated thereview. Shihua Zhu (TAR Training Fellow) andMary Kilonzo (Research Fellow) drafted the chapteron cost-effectiveness, supervised by Luke Vale(Professor of Health Technology Assessment).Shihua Zhu, TR Leyshon Griffiths (Senior Lecturerand Honorary Consultant Urological Surgeon)and Ghulam Nabi (Clinical Lecturer in Urology)drafted the background chapter. Charles Boachie(Statistician) drafted the data analysis section ofthe review and conducted the statistical analysis,supervised by Jonathan Cook (Statistician). TRLeyshon Griffiths, James N’Dow (Professor ofUrology) and Ghulam Nabi provided expert adviceon clinical aspects of the review. Cynthia Fraser(Information Officer) developed and ran the searchstrategies, obtained papers and formatted thereferences. All authors assisted in preparing themanuscript, reading and commenting on drafts,and reading the final draft.Graham Mowatt (Research Fellow) screenedthe search results, assessed full-text studies for139© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4References1. World Health Organization. Internationalclassificiation of diseases and related health problems(ICD) 10th revision. Geneva: World HealthOrganization; 2007. URL: www.who.int/classifications/apps/icd/icd10online/. Accesssed July2008.2. UroVysion fluorescence in situ hybridisation (FISH) assay.MSAC Application 1084. Canberra: Medical ServicesAdvisory Committee; 2005. URL: www.msac.gov.au/internet/msac/publishing.nsf/Content/4753418A5C8F33DDCA25745E000A3933/$File/1084%20-%20UroVysion%20Report.pdf. Accesssed September2008.3. Babjuk M, Oosterlinck W, Sylvester R, Kaasinen E,Bohle A, Palou J. Guidelines on TaT1 (non-muscleinvasive)bladder cancer [document on the internet].Arnhem: European Association of Urology; 2009[accesssed May 2009]. URL: www.uroweb.org/fileadmin/tx_eauguidelines/2009/Full/TaT1_BC.pdf.4. Kirkali Z, Chan T, Manoharan M, Algaba F, BuschC, Cheng L, et al. Bladder cancer: epidemiology,staging and grading, and diagnosis. Urology2005;66:4–34.5. Hall MC, Chang SS, Dalbagni G, Pruthi RS,Schellhammer PF, Seigne JD, et al. Guideline forthe management of nonmuscle invasive bladder cancer:(stages Ta,T1 and Tis). 2007 update. Linthicum, MD:American Urological Association; 2007. URL: www.auanet.org/guidelines/bladcan07.cfm. Accesssed July2008.6. Brennan P, Bogillot O, Cordier S, Greiser E, SchillW, Vineis P, et al. Cigarette smoking and bladdercancer in men: a pooled analysis of 11 case–controlstudies. Int J Cancer 2000;86:289–94.7. Brennan P, Bogillot O, Greiser E, Chang-ClaudeJ, Wahrendorf J, Cordier S, et al. The contributionof cigarette smoking to bladder cancer in women(pooled European data). Cancer Causes Control2001;12:411–7.8. Occupational bladder cancer: a guide for clinicians.The BAUS Subcommittee on Industrial BladderCancer. Br J Urol 1988;61:183–91.9. Steinmaus CM, Nunez S, Smith AH. Diet andbladder cancer: a meta-analysis of six dietaryvariables. Am J Epidemiol 2000;151:693–702.© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.10. Murta-Nascimento C, Schmitz-Drager BJ, ZeegersMP, Steineck G, Kogevinas M, Real FX, et al.Epidemiology of urinary bladder cancer: fromtumor development to patient’s death. World J Urol2007;25:285–95.11. Boorjian S, Cowan JE, Konety BR, DuChane J,Tewari A, Carroll PR, et al. Bladder cancer incidenceand risk factors in men with prostate cancer: resultsfrom Cancer of the Prostate Strategic UrologicResearch Endeavor. J Urol 2007;177:883–7.12. Duncan RE, Bennett DW, Evans AT, Aron BS,Schellhas HF. Radiation-induced bladder tumors. JUrol 1977;118:43–5.13. Garcia-Closas M, Malats N, Silverman D,Dosemeci M, Kogevinas M, Hein DW, et al. NAT2slow acetylation, GSTM1 null genotype, andrisk of bladder cancer: results from the SpanishBladder Cancer Study and meta-analyses. Lancet2005;366:649–59.14. Engel LS, Taioli E, Pfeiffer R, Garcia-Closas M,Marcus PM, Lan Q, et al. Pooled analysis andmeta-analysis of glutathione S-transferase M1 andbladder cancer: a HuGE review. Am J Epidemiol2002;156:95–109.15. Newling DW, Robinson MR, Smith PH, ByarD, Lockwood R, Stevens I, et al. Tryptophanmetabolites, pyridoxine (vitamin B6) and theirinfluence on the recurrence rate of superficialbladder cancer. Results of a prospective, randomisedphase III study performed by the EORTC GUGroup. EORTC Genito-Urinary Tract CancerCooperative Group. Eur Urol 1995;27:110–6.16. Sobin DH, Witteking CH, editors. TNM classificationof malignant tumours. 6th edn. New York: Wiley-Liss;2002.17. Mostofi FK. International histologic classificationof tumors. A report by the Executive Committee ofthe International Council of Societies of Pathology.Cancer 1974;33:1480–4.18. Harnden P. A critical appraisal of the classificationof urothelial tumours: time for a review ofthe evidence and a radical change? BJU Int2007;99:723–5.19. Sylvester RJ, van der Meijden AP, Lamm DL.Intravesical bacillus Calmette-Guerin reduces the141

References142risk of progression in patients with superficialbladder cancer: a meta-analysis of the publishedresults of randomized clinical trials. J Urol2002;168:1964–70.20. Sylvester RJ, van der Meijden AP, Oosterlinck W,Witjes JA, Bouffioux C, Denis L, et al. Predictingrecurrence and progression in individual patientswith stage Ta T1 bladder cancer using EORTC risktables: a combined analysis of 2596 patients fromseven EORTC trials. Eur Urol 2006;49:466–77.21. Heney NM, Ahmed S, Flanagan MJ, Frable W,Corder MP, Hafermann MD, et al. Superficialbladder cancer: progression and recurrence. J Urol1983;130:1083–6.22. Cancer Research UK. UK bladder cancer statistics.2007. URL: http://info.cancerresearchuk.org/cancerstats/types/bladder/?a=5441. Accesssed July2008.23. Hayne D, Arya M, Quinn MJ, Babb PJ, BeacockCJ, Patel HR. Current trends in bladder cancer inEngland and Wales. J Urol 2004;172:1051–5.24. National Comprehensive Cancer Network. NCCNclinical practice guidelines in oncology. Bladder cancerincluding upper tract tumours and urothelial carcinomaof the prostate. V.2.2008. Fort Washington, PA:National Comprehensive Cancer Network; 2008.URL: www.nccn.org/professionals/physician_gls/PDF/bladder.pdf. Accesssed July 2008.25. Edwards TJ, Dickinson AJ, Natale S, Gosling J,McGrath JS. A prospective analysis of the diagnosticyield resulting from the attendance of 4020 patientsat a protocol-driven haematuria clinic. BJU Int2006;97:301–5.26. Khadra MH, Pickard RS, Charlton M, Powell PH,Neal DE. A prospective analysis of 1,930 patientswith hematuria to evaluate current diagnosticpractice. J Urol 2000;163:524–7.27. Pashos CL, Botteman MF, Laskin BL, RedaelliA. Bladder cancer: epidemiology, diagnosis, andmanagement. Cancer Pract 2002;10:311–22.28. BAUS Section of Oncology, British Uro-oncologyGroup (BUG). MDT (multi-disciplinary team) guidancefor the managing of bladder cancer [document on theinternet]. London: British Association of UrologicalSurgeons; 2007 [accesssed July 2008]. URL:www.bauslibrary.co.uk/PDFS/BSONC/MDT%20guidance%20for%20bladder%20cancer.pdf.29. Hall RR, et al. Proposal for changes in cystoscopicfollow up of patients with bladder cancer andadjuvant intravesical chemotherapy. BMJ1994;308:257–60.30. Mead GM, Russell M, Clark P, Harland SJ, HarperPG, Cowan R, et al. A randomized trial comparingmethotrexate and vinblastine (MV) with cisplatin,methotrexate and vinblastine (CMV) in advancedtransitional cell carcinoma: results and a report onprognostic factors in a Medical Research Councilstudy. MRC Advanced Bladder Cancer WorkingParty. Br J Cancer 1998;78:1067–75.31. Saxman SB, Propert KJ, Einhorn LH, Crawford ED,Tannock I, Raghavan D, et al. Long-term follow-upof a phase III intergroup study of cisplatin alone orin combination with methotrexate, vinblastine, anddoxorubicin in patients with metastatic urothelialcarcinoma: a cooperative group study. J Clin Oncol1997;15:2564–9.32. Sangar VK, Ragavan N, Matanhelia SS, Watson MW,Blades RA. The economic consequences of prostateand bladder cancer in the UK. BJU Int 2005;95:59–63.33. Bosanquet N, Sikora K. The economics of cancercare in the UK. Lancet Oncol 2004;5:568–74.34. Impoving outcomes in urological cancers. Guidanceon cancer services: the manual. London: NationalInstitute for Clinical Excellence; 2002. URL: www.nice.org.uk/nicemedia/pdf/Urological_Manual.pdf.Accesssed July 2008.35. National Institute for Clinical Excellence.Laparoscopic cystectomy. Interventional proceduresguidance IPG026. London: National Institute forClinical Excellence; 2003. URL: www.nice.org.uk/nicemedia/pdf/ip/ipg026guidance.pdf. AccesssedJuly 2008.36. Scottish Intercollegiate Guidelines Network.Management of transitional cell carcinoma of thebladder. SIGN Guideline No. 85. Edinburgh: ScottishIntercollegiate Guidelines Network; 2005. URL:www.sign.ac.uk/pdf/sign85.pdf. Accesssed July 2008.37. National Institute for Health and ClinicalExcellence. Intravesical microwave hyperthermiawith intravesical chemotherapy for superficial bladdercancer. Interventional procedure guidance IPG235.London: National Institute for Health and ClinicalExcellence; 2007. URL: www.nice.org.uk/Guidance/IPG235/Guidance/pdf/English. Accesssed July 2008.38. Foster M, Barnish L, Tudway D, Guest P, JamesN, McGarr C, et al. Guidelines for the management ofbladder. Birmingham: NHS Pan-Birmingham CancerNetwork; 2007. URL: www.birminghamcancer.nhs.uk/viewdoc.ashx?id=wORYwedcX6oZ1cEsM0xp6g%3d%3d. Accesssed July 2008.39. National Screening Committee. Evaluationof urinary tract malignancy (bladder cancer)screening against NSC criteria. London: UK

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4National Screening Committee; 2002. URL:www.library.nhs.uk/screening/ViewResource.aspx?resID=60477&tabID=288. Accesssed July2008.40. Stenzl A, Cowan NC, de Santis M, Jakse G, KuczykM, Merseburger MS, et al. Guideline on bladdercancer: muscle-invasive and metastatic [document onthe internet]. Arnhem: European Association ofUrology; 2009 [accesssed July 2008]. URL: www.uroweb.org/fileadmin/tx_eauguidelines/2009/Full/Muscle-Invasive_BC.pdf.41. Lokeshwar VB, et al. Bladder tumor markers beyondcytology: International Consensus Panel on bladdertumor markers. Urology 2005;66:35–63.42. Marti A, Jichlinski P, Lange N, Ballini JP, Guillou L,Leisinger HJ, et al. Comparison of aminolevulinicacid and hexylester aminolevulinate inducedprotoporphyrin IX distribution in human bladdercancer. J Urol 2003;170:428–32.43. Grossman HB, Blute ML, Dinney CP, Jones JS,Liou LS, Reuter VE, et al. The use of urine-basedbiomarkers in bladder cancer. Urology 2006;67:62–4.44. Wazait HD, Al Bhueissi SZ, Patel HR, Nathan MS,Miller RA. Long-term surveillance of bladdertumours: current practice in the United Kingdomand Ireland. Eur Urol 2003;43:485–8.45. Chahal R, Darshane A, Browning AJ, Sundaram SK.Evaluation of the clinical value of urinary NMP22as a marker in the screening and surveillance oftransitional cell carcinoma of the urinary bladder.Eur Urol 2001;40:415–20.46. Whiting P, Rutjes AW, Reitsma JB, Bossuyt PM,Kleijnen J. The development of QUADAS: a toolfor the quality assessment of studies of diagnosticaccuracy included in systematic reviews. BMC MedRes Methodol 2003;3:25.47. Verhagen AP, de Vet HC, de Bie RA, Kessels AG,Boers M, Bouter LM, et al. The Delphi list: a criterialist for quality assessment of randomized clinicaltrials for conducting systematic reviews developedby Delphi consensus. J Clin Epidemiol 1998;51:1235–41.48. Macaskill P. Empirical Bayes estimates generatedin a hierarchical summary ROC analysis agreedclosely with those of a full Bayesian analysis. J ClinEpidemiol 2004;57:925–32.49. Rutter CM, Gatsonis CA. A hierarchical regressionapproach to meta-analysis of diagnostic testaccuracy evaluations. Stat Med 2001;20:2865–84.50. Cheng CW, Lau WK, Tan PH, Olivo M. Cystoscopicdiagnosis of bladder cancer by intravesical© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.instillation of 5-aminolevulinic acid inducedporphyrin fluorescence – the Singapore experience.Ann Acad Med Singapore 2000;29:153–8.51. Colombo R, Naspro R, Bellinzoni P, Fabbri F,Guazzoni G, Scattoni V, et al. Photodynamicdiagnosis for follow-up of carcinoma in situ of thebladder. Therap Clin Risk Manag 2007;3:1003–7.52. D’Hallewin MA, De Witte PA, Waelkens E,Merlevede W, Baert L. Fluorescence detection offlat bladder carcinoma in situ after intravesicalinstillation of hypericin. J Urol 2000;164:349–51.53. De Dominicis C, Liberti M, Perugia G, De Nunzio C,Sciobica F, Zuccala A, et al. Role of 5-aminolevulinicacid in the diagnosis and treatment of superficialbladder cancer: improvement in diagnosticsensitivity. Urology 2001;57:1059–62.54. Ehsan A, Sommer F, Haupt G, Engelmann U.Significance of fluorescence cystoscopy for diagnosisof superficial bladder cancer after intravesicalinstillation of delta aminolevulinic acid. Urol Int2001;67:298–304.55. Filbeck T, Roessler W, Knuechel R, Straub M, KielHJ, Wieland WF. 5-aminolevulinic acid-inducedfluorescence endoscopy applied at secondarytransurethral resection after conventional resectionof primary superficial bladder tumors. Urology1999;53:77–81.56. Filbeck T, Roessler W, Knuechel R, Straub M, KielHJ, Wieland WF. Clinical results of the transurethralresection and evaluation of superficial bladdercarcinomas by means of fluorescence diagnosis afterintravesical instillation of 5-aminolevulinic acid. JEndourol 1999;13:117–21.57. Fradet Y, Grossman HB, Gomella L, Lerner S,Cookson M, Albala D, et al. A comparison ofhexaminolevulinate fluorescence cystoscopyand white light cystoscopy for the detection ofcarcinoma in situ in patients with bladder cancer: aphase III, multicenter study. J Urol 2007;178:68–73.58. Frimberger D, Zaak D, Stepp H, Knuchel R,Baumgartner R, Schneede P, et al. Autofluorescenceimaging to optimize 5-ALA-induced fluorescenceendoscopy of bladder carcinoma. Urology2001;58:372–5.59. Grimbergen MC, van Swol CF, Jonges TG, BoonTA, van Moorselaar RJ. Reduced specificity of5-ALA induced fluorescence in photodynamicdiagnosis of transitional cell carcinoma afterprevious intravesical therapy. Eur Urol 2003;44:51–6.60. Hendricksen K, Moonen PM, der Heijden AG,Witjes JA. False-positive lesions detected by143

References144fluorescence cystoscopy: any association with p53and p16 expression? World J Urol 2006;24:597–601.61. Hungerhuber E, Stepp H, Kriegmair M, StiefC, Hofstetter A, Hartmann A, et al. Seven years’experience with 5-aminolevulinic acid in detectionof transitional cell carcinoma of the bladder. Urology2007;69:260–4.62. Jeon SS, Kang I, Hong JH, Choi HY, Chai SE.Diagnostic efficacy of fluorescence cystoscopyfor detection of urothelial neoplasms. J Endourol2001;15:753–9.63. Jichlinski P, Forrer M, Mizeret J, Glanzmann T,Braichotte D, Wagnieres G, et al. Clinical evaluationof a method for detecting superficial surgicaltransitional cell carcinoma of the bladder bylight-induced fluorescence of protoporphyrin IXfollowing the topical application of 5-aminolevulinicacid: preliminary results. Lasers Surg Med1997;20:402–8.64. Jichlinski P, Wagnieres G, Forrer M, Mizeret J,Guillou L, Oswald M, et al. Clinical assessment offluorescence cytoscopy during transurethral bladderresection in superficial bladder cancer. Urol Res1997;25:S3–6.65. Jichlinski P, Guillou L, Karlsen SJ, MalmstromPU, Jocham D, Brennhovd B, et al. Hexylaminolevulinate fluorescence cystoscopy: newdiagnostic tool for photodiagnosis of superficialbladder cancer – a multicenter study. J Urol2003;170:226–9.66. Jocham D, Witjes F, Wagner S, Zeylemaker B,van Moorselaar J, Grimm MO, et al. Improveddetection and treatment of bladder cancer usinghexaminolevulinate imaging: a prospective, phaseIII multicenter study. J Urol 2005;174:862–6.67. Koenig F, McGovern FJ, Larne R, Enquist H,Schomacker KT, Deutsch TF. Diagnosis of bladdercarcinoma using protoporphyrin IX fluorescenceinduced by 5-aminolaevulinic acid. BJU Int1999;83:129–35.68. Kriegmair M, Baumgartner R, Knuechel R,Steinbach P, Ehsan A, Lumper W, et al. Fluorescencephotodetection of neoplastic urothelial lesionsfollowing intravesical instillation of 5-aminolevulinicacid. Urology 1994;44:836–41.69. Kriegmair M, Stepp H, Steinbach P, Lumper W,Ehsan A, Stepp HG, et al. Fluorescence cystoscopyfollowing intravesical instillation of 5-aminolevulinicacid: a new procedure with high sensitivity fordetection of hardly visible urothelial neoplasias. UrolInt 1995;55:190–6.70. Kriegmair M, Baumgartner R, Knuchel R, SteppH, Hofstadter F, Hofstetter A. Detection of earlybladder cancer by 5-aminolevulinic acid inducedporphyrin fluorescence. J Urol 1996;155:105–9.71. Kriegmair M, Zaak D, Stepp H, Stepp H,Baumgartner R, Knuechel R, et al. Transurethralresection and surveillance of bladder cancersupported by 5-aminolevulinic acid-inducedfluorescence endoscopy. Eur Urol 1999;36:386–92.72. Landry JL, Gelet A, Bouvier R, Dubernard JM,Martin X, Colombel M. Detection of bladderdysplasia using 5-aminolaevulinic acid-inducedporphyrin fluorescence. BJU Int 2003;91:623–6.73. Riedl CR, Plas E, Pfluger H. Fluorescence detectionof bladder tumors with 5-amino-levulinic acid. JEndourol 1999;13:755–9.74. Schneeweiss S, Kriegmair M, Stepp H. Is everythingall right if nothing seems wrong? A simple methodof assessing the diagnostic value of endoscopicprocedures when a gold standard is absent. J Urol1999;161:1116–9.75. Schneeweiss S. Sensitivity analysis of the diagnosticvalue of endoscopies in cross-sectional studies in theabsence of a gold standard. Int J Technol Assess HealthCare 2000;16:834–41.76. Sim HG, Lau WK, Olivo M, Tan PH, Cheng CW.Is photodynamic diagnosis using hypericin betterthan white-light cystoscopy for detecting superficialbladder carcinoma? BJU Int 2005;95:1215–8.77. Song X, Ye Z, Zhou S, Yang W, Zhang X, Liu J, et al.The application of 5-aminolevulinic acid-inducedfluorescence for cystoscopic diagnosis and treatmentof bladder carcinoma. Photodiagnosis Photodyn Ther2007;4:39–43.78. Szygula M, Wojciechowski B, Adamek M, PietrusaA, Kawczyk-Krupka A, Cebula W, et al. Fluorescentdiagnosis of urinary bladder cancer – a comparisonof two diagnostic modalities. Photodiagnosis PhotodynTher 2004;1:23–6.79. Szygula M, Wojciechowski B, Adamek M,Kawczyk-Krupka A, Cebula W, Zieleznik W, et al.Photodynamic vs autofluorescent diagnosis ofurinary bladder using Xillix LIFE system. Phys Med2004;20:55–7.80. Tritschler S, Scharf S, Karl A, Tilki D, KnuechelR, Hartmann A, et al. Validation of the diagnosticvalue of NMP22 BladderChek test as a marker forbladder cancer by photodynamic diagnosis. Eur Urol2007;51:403–7.81. Witjes JA, Moonen PM, van der Heijden AG.Comparison of hexaminolevulinate based flexible

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4and rigid fluorescence cystoscopy with rigid whitelight cystoscopy in bladder cancer: results of aprospective Phase II study. Eur Urol 2005;47:319–22.82. Zaak D, Kriegmair M, Stepp H, Stepp H,Baumgartner R, Oberneder R, et al. Endoscopicdetection of transitional cell carcinoma with5-aminolevulinic acid: results of 1012 fluorescenceendoscopies. Urology 2001;57:690–4.83. Zaak D, Hungerhuber E, Schneede P, SteppH, Frimberger D, Corvin S, et al. Role of5-aminolevulinic acid in the detection of urothelialpremalignant lesions. Cancer 2002;95:1234–8.84. Zaak D, Stepp H, Baumgartner R, Schneede P,Waidelich R, Frimberger D, et al. Ultraviolet-excited(308 nm) autofluorescence for bladder cancerdetection. Urology 2002;60:1029–33.85. Zumbraegel A, Bichler KH, Krause FS, Feil G,Nelde HJ. The photodynamic diagnosis (PDD) forearly detection of carcinoma and dysplasia of thebladder. Adv Exp Med Biol 2003;539:61–6.86. Babjuk M, Soukup V, Petrik R, Jirsa M, DvoracekJ. 5-aminolaevulinic acid-induced fluorescencecystoscopy during transurethral resection reducesthe risk of recurrence in stage Ta/T1 bladder cancer.BJU Int 2005;96:798–802.87. Burger M, Zaak D, Stief CG, Filbeck T, WielandWF, Roessler W, et al. Photodynamic diagnosticsand noninvasive bladder cancer: is it cost-effectivein long-term application? A Germany-based costanalysis. Eur Urol 2007;52:142–7.88. Daniltchenko DI, Riedl CR, Sachs MD, KoenigF, Daha KL, Pflueger H, et al. Long-term benefitof 5-aminolevulinic acid fluorescence assistedtransurethral resection of superficial bladder cancer:5-year results of a prospective randomized study. JUrol 2005;174:2129–33.89. Denzinger S, Burger M, Walter B, Knuechel R,Roessler W, Wieland WF, et al. Clinically relevantreduction in risk of recurrence of superficialbladder cancer using 5-aminolevulinic acid-inducedfluorescence diagnosis: 8-year results of prospectiverandomized study. Urology 2007;69:675–9.90. Denzinger S, Wieland WF, Otto W, Filbeck T,Knuechel R, Burger M. Does photodynamictransurethral resection of bladder tumour improvethe outcome of initial T1 high-grade bladdercancer? A long-term follow-up of a randomizedstudy. BJU Int 2008;101:566–9.91. Filbeck T, Pichlmeier U, Knuechel R, Wieland WF,Roessler W. Clinically relevant improvement of© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.recurrence-free survival with 5-aminolevulinic acidinduced fluorescence diagnosis in patients withsuperficial bladder tumors. J Urol 2002;168:67–71.92. Kriegmair M, Zaak D, Rothenberger KH, RassweilerJ, Jocham D, Eisenberger F, et al. Transurethralresection for bladder cancer using 5-aminolevulinicacid induced fluorescence endoscopy versus whitelight endoscopy. J Urol 2002;168:475–8.93. Riedl CR, Daniltchenko D, Koenig F, Simak R,Loening SA, Pflueger H. Fluorescence endoscopywith 5-aminolevulinic acid reduces earlyrecurrence rate in superficial bladder cancer. J Urol2001;165:1121–3.94. Daniely M, Rona R, Kaplan T, Olsfanger S, ElboimL, Freiberger A, et al. Combined morphologicand fluorescence in situ hybridization analysis ofvoided urine samples for the detection and followupof bladder cancer in patients with benign urinecytology. Cancer 2007;111:517–24.95. Friedrich MG, Toma MI, Hellstern A, Pantel K,Weisenberger DJ, Noldus J, et al. Comparison ofmultitarget fluorescence in situ hybridization inurine with other noninvasive tests for detectingbladder cancer. BJU Int 2003;92:911–4.96. Friedrich MG, Hellstern A, Hautmann SH, GraefenM, Conrad S, Huland E, et al. Clinical use of urinarymarkers for the detection and prognosis of bladdercarcinoma: a comparison of immunocytologywith monoclonal antibodies against Lewis X and486p3/12 with the BTA STAT and NMP22 tests. JUrol 2002;168:470–4.97. Halling KC, King W, Sokolova IA, Meyer RG,Burkhardt HM, Halling AC, et al. A comparisonof cytology and fluorescence in situ hybridizationfor the detection of urothelial carcinoma. J Urol2000;164:1768–75.98. Junker K, Fritsch T, Hartmann A, Schulze W,Schubert J. Multicolor fluorescence in situhybridization (M-FISH) on cells from urine for thedetection of bladder cancer. Cytogenet Genome Res2006;114:279–83.99. Kipp BR, Halling KC, Campion MB, Wendel AJ,Karnes RJ, Zhang J, et al. Assessing the value ofreflex fluorescence in situ hybridization testingin the diagnosis of bladder cancer when routineurine cytological examination is equivocal. J Urol2008;179:1296–301.100. Meiers I, Singh H, Hossain D, Lang K, Liu L,Qian JQ, et al. Improved filter method for urinesediment detection of urothelial carcinoma byfluorescence in situ hybridization. Arch Pathol LabMed 2007;131:1574–7.145

References146101. Mian C, Lodde M, Comploj E, Negri G, Egarter-Vigl E, Lusuardi L, et al. Liquid-based cytology as atool for the performance of uCyt+ and UrovysionMulticolour-FISH in the detection of urothelialcarcinoma. Cytopathology 2003;14:338–42.102. Moonen PM, Merkx GF, Peelen P, KarthausHF, Smeets DF, Witjes JA. UroVysion comparedwith cytology and quantitative cytology in thesurveillance of non-muscle invasive bladder cancer.Eur Urol 2007;51:1275–80.103. Sarosdy MF, Kahn PR, Ziffer MD, Love WR,Barkin J, Abara EO, et al. Use of a multitargetfluorescence in situ hybridization assay to diagnosebladder cancer in patients with hematuria. J Urol2006;176:44–7.104. Skacel M, Fahmy M, Brainard JA, Pettay JD,Biscotti CV, Liou LS, et al. Multitarget fluorescencein situ hybridization assay detects transitional cellcarcinoma in the majority of patients with bladdercancer and atypical or negative urine cytology. JUrol 2003;169:2101–5.105. Sokolova IA, Halling KC, Jenkins RB, BurkhardtHM, Meyer RG, Seelig SA, et al. The developmentof a multitarget, multicolor fluorescence in situhybridization assay for the detection of urothelialcarcinoma in urine. J Mol Diagn 2000;2:116–23.106. Yoder BJ, Skacel M, Hedgepeth R, Babineau D,Ulchaker JC, Liou LS, et al. Reflex UroVysiontesting of bladder cancer surveillance patientswith equivocal or negative urine cytology: aprospective study with focus on the natural historyof anticipatory positive findings. Am J Clin Pathol2007;127:295–301.107. May M, Hakenberg OW, Gunia S, Pohling P, HelkeC, Lubbe L, et al. Comparative diagnostic value ofurine cytology, UBC-ELISA, and fluorescence insitu hybridization for detection of transitional cellcarcinoma of urinary bladder in routine clinicalpractice. Urology 2007;70:449–53.108. Sarosdy MF, Schellhammer P, Bokinsky G, Kahn P,Chao R, Yore L, et al. Clinical evaluation of a multitargetfluorescent in situ hybridization assay fordetection of bladder cancer. J Urol 2002;168:1950–4.109. Lodde M, Mian C, Negri G, Berner L, Maffei N,Lusuardi L, et al. Role of uCyt+ in the detectionand surveillance of urothelial carcinoma. Urology2003;61:243–7.110. Lodde M, Mian C, Comploj E, Palermo S, LonghiE, Marberger M, et al. uCyt+ test: alternative tocystoscopy for less-invasive follow-up of patientswith low risk of urothelial carcinoma. Urology2006;67:950–4.111. Messing EM, Teot L, Korman H, Underhill E,Barker E, Stork B, et al. Performance of urine testin patients monitored for recurrence of bladdercancer: a multicenter study in the United States. JUrol 2005;174:1238–41.112. Mian C, Pycha A, Wiener H, Haitel A, Lodde M,Marberger M. Immunocyt: a new tool for detectingtransitional cell cancer of the urinary tract. J Urol1999;161:1486–9.113. Mian C, Maier K, Comploj E, Lodde M, BernerL, Lusuardi L, et al. uCyt+/ImmunoCyt in thedetection of recurrent urothelial carcinoma: anupdate on 1991 analyses. Cancer 2006;108:60–5.114. Olsson H, Zackrisson B. ImmunoCyt a usefulmethod in the follow-up protocol for patients withurinary bladder carcinoma. Scand J Urol Nephrol2001;35:280–2.115. Pfister C, Chautard D, Devonec M, Perrin P, ChopinD, Rischmann P, et al. Immunocyt test improves thediagnostic accuracy of urinary cytology: results of aFrench multicenter study. J Urol 2003;169:921–4.116. Piaton E, Daniel L, Verriele V, Dalifard I,Zimmermann U, Renaudin K, et al. Improveddetection of urothelial carcinomas with fluorescenceimmunocytochemistry (uCyt+ assay) and urinarycytology: results of a French Prospective MulticenterStudy. Lab Invest 2003;83:845–52.117. Schmitz-Drager BJ, Beiche B, Tirsar LA, Schmitz-Drager C, Bismarck E, Ebert T. Immunocytologyin the assessment of patients with asymptomaticmicrohaematuria. Eur Urol 2007;51:1582–8.118. Schmitz-Drager BJ, Tirsar LA, Schmitz-DragerC, Dorsam J, Mellan Z, Bismarck E, et al.Immunocytology in the assessment of patients withasymptomatic hematuria. World J Urol 2008;26:31–7.119. Tetu B, Tiguert R, Harel F, Fradet Y. ImmunoCyt/uCyt+ improves the sensitivity of urine cytologyin patients followed for urothelial carcinoma. ModPathol 2005;18:83–9.120. Bhuiyan J, Akhter J, O’Kane DJ. Performancecharacteristics of multiple urinary tumor markersand sample collection techniques in the detection oftransitional cell carcinoma of the bladder. Clin ChimActa 2003;331:69–77.121. Casella R, Huber P, Blochlinger A, Stoffel F,Dalquen P, Gasser TC, et al. Urinary level of nuclearmatrix protein 22 in the diagnosis of bladdercancer: experience with 130 patients with biopsyconfirmed tumor. J Urol 2000;164:1926–8.

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4122. Casetta G, Gontero P, Zitella A, Pelucelli G,Formiconi A, Priolo G, et al. BTA quantitative assayand NMP22 testing compared with urine cytologyin the detection of transitional cell carcinoma of thebladder. Urol Int 2000;65:100–5.123. Del Nero A, Esposito N, Curro A, Biasoni D,Montanari E, Mangiarotti B, et al. Evaluation ofurinary level of NMP22 as a diagnostic markerfor stage pTa-pT1 bladder cancer: comparisonwith urinary cytology and BTA test. Eur Urol1999;35:93–7.124. Giannopoulos A, Manousakas T, Mitropoulos D,Botsoli-Stergiou E, Constantinides C, GiannopoulouM, et al. Comparative evaluation of the BTA stattest, NMP22, and voided urine cytology in thedetection of primary and recurrent bladder tumors.Urology 2000;55:871–5.125. Giannopoulos A, Manousakas T, Gounari A,Constantinides C, Choremi-Papadopoulou H,Dimopoulos C. Comparative evaluation of thediagnostic performance of the BTA stat test, NMP22and urinary bladder cancer antigen for primaryand recurrent bladder tumors [see comment]. J Urol2001;166:470–5.126. Grossman HB, Messing E, Soloway M, TomeraK, Katz G, Berger Y, et al. Detection of bladdercancer using a point-of-care proteomic assay. JAMA2005;293:810–6.127. Grossman HB, Soloway M, Messing E, Katz G, SteinB, Kassabian V, et al. Surveillance for recurrentbladder cancer using a point-of-care proteomicassay. JAMA 2006;295:299–305.128. Gutierrez Banos JL, Rebollo Rodrigo MH, AntolinJuarez FM, Martin GB. NMP 22, BTA stat testand cytology in the diagnosis of bladder cancer: acomparative study. Urol Int 2001;66:185–90.129. Hughes JH, Katz RL, Rodriguez-Villanueva J, KiddL, Dinney C, Grossman HB, et al. Urinary nuclearmatrix protein 22 (NMP22): a diagnostic adjunctto urine cytologic examination for the detection ofrecurrent transitional-cell carcinoma of the bladder.Diagn Cytopathol 1999;20:285–90.130. Kowalska M, Kaminska J, Kotowicz B, FuksiewiczM, Rysinska A, Demkow T, et al. Evaluation ofthe urinary nuclear matrix protein (NMP22)as a tumour marker in bladder cancer patients.Nowotwory 2005;55:300–2.131. Kumar A, Kumar R, Gupta NP. Comparison ofNMP22 BladderChek test and urine cytology forthe detection of recurrent bladder cancer. Jpn J ClinOncol 2006;36:172–5.© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.132. Lahme S, Bichler KH, Feil G, Krause S. Comparisonof cytology and nuclear matrix protein 22 for thedetection and follow-up of bladder cancer. Urol Int2001;66:72–7.133. Lahme S, Bichler KH, Feil G, Zumbragel A, Gotz T.Comparison of cytology and nuclear matrix protein22 (NMP 22) for the detection and follow-up ofbladder-cancer. Adv Exp Med Biol 2003;539:111–9.134. Mian C, Lodde M, Haitel A, Vigl EE, MarbergerM, Pycha A. Comparison of the monoclonal UBC-ELISA test and the NMP22 ELISA test for thedetection of urothelial cell carcinoma of the bladder.Urology 2000;55:223–6.135. Miyanaga N, Akaza H, Tsukamoto T, Ishikawa S,Noguchi R, Ohtani M, et al. Urinary nuclear matrixprotein 22 as a new marker for the screening ofurothelial cancer in patients with microscopichematuria. Int J Urol 1999;6:173–7.136. Miyanaga N, Akaza H, Tsukamoto S, Shimazui T,Ohtani M, Ishikawa S, et al. Usefulness of urinaryNMP22 to detect tumor recurrence of superficialbladder cancer after transurethral resection. Int JClin Oncol 2003;8:369–73.137. Oge O, Atsu N, Kendi S, Ozen H. Evaluation ofnuclear matrix protein 22 (NMP22) as a tumormarker in the detection of bladder cancer. Int UrolNephrol 2001;32:367–70.138. Oosterhuis JWA, Pauwels RPE, Schapers RFM,Van Pelt J, Smeets W, Newling DWW. Detection ofrecurrent transitional cell carcinoma of the bladderwith nuclear matrix protein-22 in a follow-upsetting. Urooncology 2002;2:137–42.139. Ponsky LE, Sharma S, Pandrangi L, Kedia S, NelsonD, Agarwal A, et al. Screening and monitoring forbladder cancer: refining the use of NMP22. J Urol2001;166:75–8.140. Poulakis V, Witzsch U, De Vries R, AltmannsbergerHM, Manyak MJ, Becht E. A comparison ofurinary nuclear matrix protein-22 and bladdertumour antigen tests with voided urinary cytologyin detecting and following bladder cancer: theprognostic value of false-positive results. BJU Int2001;88:692–701.141. Saad A, Hanbury DC, McNicholas TA, BousteadGB, Morgan S, Woodman AC. A study comparingvarious noninvasive methods of detecting bladdercancer in urine. BJU Int 2002;89:369–73.142. Sanchez-Carbayo M, Urrutia M, Gonzalez deBuitrago JM, Navajo JA. Utility of serial urinarytumor markers to individualize intervals betweencystoscopies in the monitoring of patients withbladder carcinoma. Cancer 2001;92:2820–8.147

References148143. Serretta V, Lo PD, Vasile P, Gange E, EspositoE, Menozzi I. Urinary NMP22 for the detectionof recurrence after transurethral resection oftransitional cell carcinoma of the bladder:experience on 137 patients. Urology 1998;52:793–6.144. Serretta V, Pomara G, Rizzo I, Esposito E. UrinaryBTA-stat, BTA-trak and NMP22 in surveillanceafter TUR of recurrent superficial transitional cellcarcinoma of the bladder. Eur Urol 2000;38:419–25.145. Shariat SF, Casella R, Monoski MA, Sulser T, GasserTC, Lerner SP. The addition of urinary urokinasetypeplasminogen activator to urinary nuclearmatrix protein 22 and cytology improves thedetection of bladder cancer. J Urol 2003;170:2244–7.146. Shariat SF, Casella R, Khoddami SM, HernandezG, Sulser T, Gasser TC, et al. Urine detection ofsurvivin is a sensitive marker for the noninvasivediagnosis of bladder cancer. J Urol 2004;171:626–30.147. Shariat SF, Marberger MJ, Lotan Y, Sanchez-Carbayo M, Zippe C, Ludecke G, et al. Variabilityin the performance of nuclear matrix protein22 for the detection of bladder cancer. J Urol2006;176:919–26.148. Sharma S, Zippe CD, Pandrangi L, Nelson D,Agarwal A. Exclusion criteria enhance the specificityand positive predictive value of NMP22 and BTAstat. J Urol 1999;162:53–7.149. Stampfer DS, Carpinito GA, Rodriguez-VillanuevaJ, Willsey LW, Dinney CP, Grossman HB, etal. Evaluation of NMP22 in the detection oftransitional cell carcinoma of the bladder. J Urol1998;159:394–8.150. Talwar R, Sinha T, Karan SC, Doddamani D,Sandhu A, Sethi GS, et al. Voided urinary cytology inbladder cancer: is it time to review the indications?Urology 2007;70:267–71.151. Wiener HG, Mian C, Haitel A, Pycha A, Schatzl G,Marberger M. Can urine bound diagnostic testsreplace cystoscopy in the management of bladdercancer? J Urol 1998;159:1876–80.152. Zippe C, Pandrangi L, Potts JM, Kursh E, NovickA, Agarwal A. NMP22: a sensitive, cost-effective testin patients at risk for bladder cancer. Anticancer Res1999;19:2621–3.153. Zippe C, Pandrangi L, Agarwal A. NMP22 is asensitive, cost-effective test in patients at risk forbladder cancer. J Urol 1999;161:62–5.154. Abbate I, D’Introno A, Cardo G, Marano A,Addabbo L, Musci MD, et al. Comparison of nuclearmatrix protein 22 and bladder tumor antigen inurine of patients with bladder cancer. Anticancer Res1998;18:3803–5.155. Boman H, Hedelin H, Holmang S. Four bladdertumor markers have a disappointingly lowsensitivity for small size and low grade recurrence. JUrol 2002;167:80–3.156. Chang YH, Wu CH, Lee YL, Huang PH, Kao YL,Shiau MY. Evaluation of nuclear matrix protein-22as a clinical diagnostic marker for bladder cancer.Urology 2004;64:687–92.157. Lee KH. Evaluation of the NMP22 test andcomparison with voided urine cytology inthe detection of bladder cancer. Yonsei Med J2001;42:14–8.158. Parekattil SJ, Fisher HA, Kogan BA. Neural networkusing combined urine nuclear matrix protein-22,monocyte chemoattractant protein-1 and urinaryintercellular adhesion molecule-1 to detect bladdercancer. J Urol 2003;169:917–20.159. Ramakumar S, Bhuiyan J, Besse JA, RobertsSG, Wollan PC, Blute ML, et al. Comparison ofscreening methods in the detection of bladdercancer. J Urol 1999;161:388–94.160. Sanchez-Carbayo M, Herrero E, Megias J, Mira A,Soria F. Evaluation of nuclear matrix protein 22 asa tumour marker in the detection of transitional cellcarcinoma of the bladder. BJU Int 1999;84:706–13.161. Sanchez-Carbayo M, Herrero E, Megias J, MiraA, Soria F. Comparative sensitivity of urinaryCYFRA 21–1, urinary bladder cancer antigen,tissue polypeptide antigen, tissue polypeptideantigen and NMP22 to detect bladder cancer. J Urol1999;162:1951–6.162. Sanchez-Carbayo M, Urrutia M, Silva JM, RomaniR, De Buitrago JM, Navajo JA. Comparativepredictive values of urinary cytology, urinarybladder cancer antigen, CYFRA 21–1 and NMP22for evaluating symptomatic patients at risk forbladder cancer. J Urol 2001;165:1462–7.163. Sozen S, Biri H, Sinik Z, Kupeli B, Alkibay T,Bozkirli I. Comparison of the nuclear matrixprotein 22 with voided urine cytology and BTA stattest in the diagnosis of transitional cell carcinoma ofthe bladder. Eur Urol 1999;36:225–9.164. Takeuchi Y, Sawada Y. A clinical study of urinaryNMP22 in urinary epithelial cancer. J Med Soc TohoUniv 2004;51:332–8.165. Bastacky S, Ibrahim S, Wilczynski SP, Murphy WM.The accuracy of urinary cytology in daily practice.Cancer 1999;87:118–28.

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4166. Chahal R, Gogoi NK, Sundaram SK. Is it necessaryto perform urine cytology in screening patients withhaematuria? Eur Urol 2001;39:283–6.167. Garbar C, Mascaux C, Wespes E. Is urinary tractcytology still useful for diagnosis of bladdercarcinomas? A large series of 592 bladder washingsusing a five-category classification of differentcytological diagnoses. Cytopathology 2007;18:79–83.168. Hakenberg OW, Franke P, Froehner M, ManseckA, Wirth MP. The value of conventional urinecytology in the diagnosis of residual tumour aftertransurethral resection of bladder carcinomas.Onkologie 2000;23:252–7.169. Hutterer GC, Karakiewicz PI, Zippe C, LudeckeG, Boman H, Sanchez-Carbayo M, et al. Urinarycytology and nuclear matrix protein 22 in thedetection of bladder cancer recurrence other thantransitional cell carcinoma. BJU Int 2008;101:561–5.170. Karakiewicz PI, Benayoun S, Zippe C, Ludecke G,Boman H, Sanchez-Carbayo M, et al. Institutionalvariability in the accuracy of urinary cytology forpredicting recurrence of transitional cell carcinomaof the bladder. BJU Int 2006;97:997–1001.171. Planz B, Jochims E, Deix T, Caspers HP, JakseG, Boecking A. The role of urinary cytology fordetection of bladder cancer. Eur J Surg Oncol2005;31:304–8.172. Potter JM, Quigley M, Pengelly AW, Fawcett DP,Malone PR. The role of urine cytology in theassessment of lower urinary tract symptoms. BJU Int1999;84:30–1.173. Raitanen M-P, Aine R, Rintala E, Kallio J, RajalaP, Juusela H, et al. Differences between local andreview urinary cytology in diagnosis of bladdercancer. An interobserver multicenter analysis. EurUrol 2002;41:284–9.174. Raitanen MP, Aine RA, Kaasinen ES, LiukkonenTJ, Kylmala TM, Huhtala H, et al. Suspicious urinecytology (class III) in patients with bladder cancer:should it be considered as negative or positive?Scand J Urol Nephrol 2002;36:213–7.175. Botteman MF, Pashos CL, Redaelli A, Laskin B,Hauser R. The health economics of bladder cancer:a comprehensive review of the published literature.Pharmacoeconomics 2003;21:1315–30.176. Corwin HL, Silverstein MD. The diagnosisof neoplasia in patients with asymptomaticmicroscopic hematuria: a decision analysis. J Urol1988;139:1002–6.© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.177. Ellwein LB, Farrow GM. Urinary cytology screening:the decision facing the asymptomatic patient. MedDecis Making 1988;8:110–9.178. Novicki DE, Stern JA, Nemec R, Lidner TK. Costeffectiveevaluation of indeterminate urinarycytology. J Urol 1998;160:734–6.179. Rodgers M, Nixon J, Hempel S, Aho T, Kelly J, NealD, et al. Diagnostic tests and algorithms used inthe investigation of haematuria: systematic reviewsand economic evaluation. Health Technol Assess2006;10(18).180. Lotan Y, Svatek RS, Sagalowsky AI. Should wescreen for bladder cancer in a high-risk population?A cost per life-year saved analysis. Cancer2006;107:982–90.181. Kurth KH, Denis L, Bouffioux C, Sylvester R,Debruyne FMJ, Pavone-Macaluso M, et al. Factorsaffecting recurrence and progression in superficialbladder tumours. Eur J Cancer 1995;31:1840–6.182. Dalesio O, Schulman CC, Sylvester R, De Pauw M,Robinson M, Denis L, et al. Prognostic factors insuperficial bladder tumors. A study of the EuropeanOrganization for Research on Treatment of Cancer:Genitourinary Tract Cancer Cooperative Group. JUrol 1983;129:730–3.183. García RJ, Fernandez Gomez JM, Escaf BS, JalonMA, Alvarez MM, Regadera SJ. [Pronostic factorson recurrence and progression of superficialbladder cancer. Risk groups (part II)]. Actas Urol Esp2006;30:1009–16.184. Kiemeney LA, Witjes JA, Heijbroek RP, VerbeekAL, Debruyne FM. Predictability of recurrentand progressive disease in individual patientswith primary superficial bladder cancer. J Urol1993;150:60–4.185. Kiemeney LA, Witjes JA, Heijbroek RP, DebruyneFM, Verbeek AL. Dysplasia in normal-lookingurothelium increases the risk of tumour progressionin primary superficial bladder cancer. Eur J Cancer1994;30A:1621–5.186. Loening S, Narayana A, Yoder L, Slymen D, PenickG, Culp D. Analysis of bladder tumor recurrence in178 patients. Urology 1980;16:137–41.187. Millán-Rodriguez F, Chéchile-Toniolo G, Salvador-Bayarri J, Palou J, Algaba F, Vicente-RodriguezJ. Primary superficial bladder cancer risk groupsaccording to progression, mortality and recurrence.J Urol 2000;164:680–4.188. Mulders PF, Meyden AP, Doesburg WH, OosterhofGO, Debruyne FM. Prognostic factors in pTa-pT1superficial bladder tumours treated with intravesical149

References150instillations. The Dutch South-Eastern UrologicalCollaborative Group. Br J Urol 1994;73:403–8.189. Narayana AS, Loening SA, Slymen DJ, Culp DA.Bladder cancer: factors affecting survival. J Urol1983;130:56–60.190. Oosterlinck W, Lobel B, Jakse G, MalmstromPU, Stockle M, Sternberg CO. Guidelines onbladder cancer. Arnhem: European Association ofUrology; 2001. URL: www.uroweb.org/fileadmin/tx_eauguidelines/2001_Bladder%20_Cancer.PDF.Accesssed September 2008.191. Parmar H, Charlton C, Phillips RH, Lightman SL.Management of advanced prostatic cancer. Lancet1989;2:1338–9.192. Pawinski A, Sylvester R, Kurth KH, BouffiouxC, Van der MA, Parmar MK, et al. A combinedanalysis of European Organization for Research andTreatment of Cancer, and Medical Research Councilrandomized clinical trials for the prophylactictreatment of stage TaT1 bladder cancer. EuropeanOrganization for Research and Treatment of CancerGenitourinary Tract Cancer Cooperative Group andthe Medical Research Council Working Party onSuperficial Bladder Cancer. J Urol 1996;156:1934–40.193. Shinka T, Matsumoto M, Ogura H, Hirano A,Ohkawa T. Recurrence of primary superficialbladder cancer treated with prophylactic intravesicalTokyo 172 bacillus Calmette-Guerin: a long-termfollow-up. Int J Urol 1997;4:139–43.194. Witjes JA, Kiemeney LALM, Verbeek ALM,Heijbroek RP, Debruyne FMJ. Random bladderbiopsies and the risk of recurrent superficial bladdercancer: a prospective study in 1026 patients. World JUrol 1992;10:231–4.195. Witjes JA, Kiemeney LA, Schaafsma HE, DebruynFM. The influence of review pathology on studyoutcome of a randomized multicentre superficialbladder cancer trial. Members of the Dutch SouthEast Cooperative Urological Group. Br J Urol1994;73:172–6.196. Herr HW. Natural history of superficial bladdertumors: 10- to 20-year follow-up of treated patients.World J Urol 1997;15:84–8.197. Stein JP, Lieskovsky G, Cote R, Groshen S, FengA-C, Boyd S, et al. Radical cystectomy in thetreatment of invasive bladder cancer: long-termresults in 1,054 patients. J Clin Oncol 2001;19:666–75.198. von der Maase H, Sengelov L, Roberts JT,Ricci S, Dogliotti L, Oliver T, et al. Long-termsurvival results of a randomized trial comparinggemcitabine plus cisplatin, with methotrexate,vinblastine, doxorubicin, plus cisplatin in patientswith bladder cancer. J Clin Oncol 2005;23:4602–8.199. Government Actuary’s Department. Interim lifetables 2004–6. London: Government Actuary’sDepartment; 2008. URL: www.gad.gov.uk/Demography%5FData/Life_Tables/Interim_life_tables.asp. Accesssed September 2008.200. Farrow SC, Fowkes FGR, Lunn JN. Epidemiologyin anaesthesia. II. Factors affecting mortality inhospital. Br J Anaesth 1982;54:811–7.201. Kondas J, Szentgyorgyi E. Transurethral resection of1250 bladder tumours. Int Urol Nephrol 1992;24:35–42.202. Department of Health. NHS reference costs2005–6. 2006. URL: www.dh.gov.uk/en/Publicationsandstatistics/Publications/PublicationsPolicyAndGuidance/DH_062884.Accesssed September 2008.203. Tumour-marker NMP22 (NMP). Nottingham:MediChecks.com; 2008. URL: http://medichecks.com/index.cfm?s=2&d=21&test=NMP. AccesssedSeptember 2008.204. National Institute for Health and ClinicalExcellence. Guide to the methods of technology appraisal.London: National Institute for Health and ClinicalExcellence; 2008. URL: www.nice.org.uk/media/B52/A7/TAMethodsGuideUpdatedJune2008.pdf.Accesssed September 2008.205. Kulkarni GS, Finelli A, Fleshner NE, JewettMAS, Lopushinsky SR, Alibhai SMH. Optimalmanagement of high-risk T1G3 bladder cancer: adecision analysis. PLoS Med 2007;4:1538–49.206. Pickard AS, Neary MP, Cella D. Estimation ofminimally important differences in EQ-5D utilityand VAS scores in cancer. Health Qual Life Outcomes2007;5:70.207. Schmidbauer J, Witjes F, Schmeller N, DonatR, Susani M, Marberger M, et al. Improveddetection of urothelial carcinoma in situ withhexaminolevulinate fluorescence cystoscopy. J Urol2004;171:135–8.208. Spiess PE, Grossman HB. Fluorescence cystoscopy:is it ready for use in routine clinical practice? CurrOpin Urol 2006;16:372–6.209. Zaak D, Karl A, Knuchel R, Stepp H, Hartmann A,Reich O, et al. Diagnosis of urothelial carcinoma ofthe bladder using fluorescence endoscopy. BJU Int2005;96:217–22.

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4210. Jain S, Kockelbergh RC. The role of photodynamicdiagnosis in the contemporary management ofsuperficial bladder cancer. BJU Int 2005;96:17–21.211. Lotan Y, Roehrborn CG. Sensitivity and specificityof commonly available bladder tumor markersversus cytology: results of a comprehensiveliterature review and meta-analyses. Urology2003;61:109–18.212. Glas AS, Roos D, Deutekom M, Zwinderman AH,Bossuyt PM, Kurth KH. Tumor markers in thediagnosis of primary bladder cancer. A systematicreview. J Urol 2003;169:1975–82.213. van Rhijn BW, van der Poel HG, van der Kwast TH.Urine markers for bladder cancer surveillance: asystematic review. Eur Urol 2005;47:736–48.214. Yossepowitch O, Herr HW, Donat SM. Use ofurinary biomarkers for bladder cancer surveillance:patient perspectives. J Urol 2007;177:1277–82.215. Kiemeney LALM, Witjes JA, Heijbroek RP, KoperNP, Verbeek ALM, Debruyne FMJ. Should randomurothelial biopsies be taken from patients withprimary superficial bladder cancer? A decisionanalysis. Br J Urol 1994;73:164–71.216. van der Meijden A, Oosterlinck W, Brausi M, KurthK-H, Sylvester R, de Balincourt C. Significanceof bladder biopsies in Ta,T1 bladder tumors: areport from the EORTC genito-urinary tract cancercooperative group. Eur Urol 1999;35:267–71.151© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Appendix 1Search strategiesClinical effectivenessMEDLINE (1966 to March Week 32008), EMBASE (1980 to 2008 Week 13),Medline In-Process (31 March 2008)Ovid Multifile SearchURL: http://gateway.ovid.com/athens1. urinary bladder neoplasms/use mesz2. exp bladder cancer/use emez3. hematuria/4. (bladder adj3 (cancer$or neoplasms$orcarci$)).tw.5. (hematuria or haematuria).tw.6. or/1–57. *urinary bladder neoplasms/su use mesz8. exp *bladder cancer/su use emez9. cystectomy/10. ((bladder adj3 resect$) or cystectomy or turbt).tw.11. or/7–1012. cystoscopy/13. cystoscop$.tw.14. (photo dynamic$or photodynamic$orfluorescence$).tw.15. (12 or 13) and 1416. hypericin.tw.17. 548–04–9.rn.18. hexvix.tw.19. hexaminolevulinate.tw.20. (hexyl$adj3 aminolevulinate).tw.21. 106–60–5.rn.22. 5-ALA.tw.23. 5-aminolevulinic acid.tw.24. 5-aminolevulinic acid hexyl ester.tw,rn.25. or/15–2426. (6 or 11) and 2527. tumor markers,biological/use mesz28. exp tumor marker/or biological marker/ordisease marker/use emez29. ((tumo?r or biological or molecular orhistolog$or biochem$or genetic$or urine ordisease) adj3 marker$).tw.30. 6 and (27 or 28 or 29))31. In Situ Hybridization, Fluorescence/32. fluorescence in situ hybridization.tw.33. urovysion.tw34. or/31–3335. 6 and 3436. nuclear proteins/© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.37. (nuclear matrix protein 22 or nmp22).tw,rn.38. or/36–3739. 6 and 3840. urine/cy41. urine cytology/use emez42. cytodiagnosis/use mesz43. cancer cytodiagnosis/use emez44. cell count/45. immunocyt$or ucyt$.tw.46. or/40–4547. 6 and 4648. 26 or 30 or 35 or 39 or 4749. (animals/or nonhuman/) not humans/50. 48 not 4951. (editorial or letter or comment or case reports).pt.52. editorial/or letter/or note/or case report/useemez53. 50 not (51 or 52)54. “sensitivity and specificity”/55. roc curve/56. receiver operating characteristic/use emez57. predictive value of tests/58. diagnostic errors/use emez59. false positive reactions/use mesz60. false negative reactions/use mesz61. diagnostic accuracy/use emez62. diagnostic value/use emez63. du.fs. use mesz64. sensitivity.tw.65. distinguish$.tw.66. differentiate.tw.67. identif$.tw.68. detect$.tw.69. diagnos$.tw.70. (predictive adj4 value$).tw.71. accura$.tw.72. comparison.tw.73. or/54–7274. 53 and 7375. exp diagnostic errors/76. reproducibility of results/77. observer variation78. exp reliability/79. diagnosis, differential/80. early diagnosis/81. (reliab$or reproduc$).tw.82. or/75–8183. 53 and 82153

Appendix 115484. prognosis/85. (predict$or prognosis or prognostic).tw.86. 84 or 8587. 53 and 8688. 26 or 74 or 83 or 87Science Citation Index (1970 to 1 April2006), BIOSIS (1985 to 3 April 2008)Web of KnowledgeURL: http://wok.mimas.ac.uk/#1 TS=(bladder SAME (cancer* or neoplasm* orcarci*))#2 TS=(hematuria OR haematuria)#3 #1 or #2#4 TS=((bladder SAME resect*) or cystectomy orturbt)#5 #3 or #4#6 TS=(cystoscop* AND (photo* dynamic* ORphotodynamic* OR fluorescence*))#7 #5 AND #6#8 TS=(hypericin or hexvix orhexaminolevulin*or hexyl* aminolevulin* or5-ala or 5-aminolevulin*)#9 #5 and #8#10 #7 or #9#11 TS=(marker* SAME (tumor or tumouror biological or molecular or histolog* orbiochem* or genetic* or urine or disease))#12 #3 and #11#13 TS=(immunocyt* or ucyt*)#14 TS=cytolog*#15 TS=(nmp22 or nuclear matrix protein 22)#16 TS=urovysion#17 TS=(fluorescence SAME hybridization)#18 #13 or #14 or #15 or #16 or #17#19 #3 and #18#20 #10 or #12 or #19#21 TS=((bladder or hemauturia or haematuria)SAME (predict* or prognosis or prognostic orreliab* or reproduc*))#22 TS=((bladder or hemauturia or haematuria)SAME (sensitivity or specificity or roc))#23 TS=((bladder or hemauturia or haematuria)SAME (identif* or accura* or compara*))#24 TS=((bladder or hemauturia or haematuria)SAME detect*)#25 TS=((bladder or hemauturia or haematuria)SAME diagnos*)#26 #21 or #22 or #23 or #24 or #25#27 #20 and #26#28 #10 or #27Health Management InformationConsortium (1979 to March 2008)Ovid Multifile SearchURL: http://gateway.ovid.com/athens1. bladder cancer/2. haematuria/3. 1 or 24. (photo$dynamic$or photodynamic orfluorescence).tw. (17)5. (hypericin or hexvix or hexyl$or 5-ala$oraminolevulonate).tw.6. (marker$or biomarker$).tw.7. (nmp22 or immunocyt$or ucyt$or urovysion orfish).tw. (8. cytology/9. or/4–810. 3 and 9Cochrane Library (Issue 1 2008)URL: http://www3.interscience.wiley.com/cgi-bin/mrwhome/106568753/HOME#1 URINARY BLADDER NEOPLASMS singleterm (MeSH)#2 HEMATURIA single term (MeSH)#3 (#1 or #2)#4 ((photo* next dynamic*) or photodynamic*or fluoresence*)#5 (hypericin or hexvix or hexyl* or ala)#6 (#4 or #5)#7 (#3 and #6)#8 marker*#9 #9 nmp22 or immunocyt or urovysion or fish#10 (#3 and (#8 or #9))#11 (#7 or #10)DARE and HTA databases (March 2008)NHS Centre for Reviews and DisseminationURL: http://nhscrd.york.ac.uk/welcome.htm1 MeSH Bladder Neoplasms EXPLODE 1 2 3 4 49# 2 MeSH Hematuria EXPLODE 1 2 15# 4 nmp22 OR immunocyt OR ucyt OR urovysionOR fish 93# 5 marker* or biomarker* 419# 7 #1 or #2 63# 8 #5 and #7 11# 9 photo AND dynamic OR photodynamic 83# 10 #7 and #9 2# 12 fluorescence OR hexvix OR hexyl ORhypericin OR 5-ala 34# 13 #1 or #2 or #4 or #8 or #10 or #12 171Medion (March 2008)URL: www.mediondatabase.nl/

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Bladder or hematuria or haematuriaNational Research Register Archive(September 2007)URL: www.update-software.com/National/#1 URINARY BLADDER NEOPLASMS singleterm (MeSH)#2 HEMATURIA single term (MeSH)#3 (#1 or #2)#4 ((photo* next dynamic*) or photodynamic*or fluoresence*)#5 (hypericin or hexvix or hexyl* or ala)#6 (#4 or #5)#7 (#3 and #6)#8 marker*#9 #9 nmp22 or immunocyt or urovysion or fish#10 (#3 and (#8 or #9))#11 (#7 or #10)ClinicalTrials.gov (March 2008)URL: http://clinicaltrials.gov/ct/gui/c/r“bladder cancer”:Topic AND (photodynamic ORfluoresence OR ALA OR hexvix OR hexyl ORhypericin or NMP22 or Immunocyt or urovysion orfish): Search termsCurrent Controlled Trials (March 2008)URL: www.controlled-trials.com/bladder AND (marker% OR photo% ORfluoresence OR ALA OR hexvix OR hexyl ORhypericin or NMP22 or Immunocyt or urovysion orfish)WHO ICTRP (March 2008)URL: www.who.int/ictrp/en/(photodynamic OR fluoresence OR ALA ORhexvix OR hexyl OR hypericin or NMP22 orImmunocyt or urovysion or fish):TI AND bladdercancer:ConditionCost-effectivenessMEDLINE (1966 to March Week 32008), EMBASE (1980 to 2008 Week 13),Medline In-Process (1 April 2008)Ovid Multifile SearchURL: http://gateway.ovid.com/athens1. urinary bladder neoplasms/use mesz2. exp bladder cancer/use emez3. hematuria/4. (bladder adj3 (cancer$or neoplasm$or carci$)).tw.© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.5. (hematuria or haematuria).tw.6. or/1–57. *urinary bladder neoplasms/su use mesz8. exp *bladder cancer/su use emez9. cystectomy/10. ((bladder adj3 resect$) or cystectomy or turbt).tw.11. or/7–1012. cystoscopy/13. cystoscop$.tw.14. (photo dynamic$or photodynamic$orfluorescence$).tw.15. (12 or 13) and 1416. hypericin.tw.17. 548–04–9.rn.18. hexvix.tw.19. hexaminolevulinate.tw.20. (hexyl$adj3 aminolevulinate).tw.21. 106–60–5.rn.22. 5-ALA.tw.23. 5-aminolevulinic acid.tw.24. 5-aminolevulinic acid hexyl ester.tw,rn.25. or/15–2426. (6 or 11) and 2527. tumor markers,biological/use mesz28. exp tumor marker/or biological marker/ordisease marker/use emez29. ((tumo?r or biological or molecular orhistolog$or biochem$or genetic$or urine ordisease) adj3 marker$).tw.30. 6 and (27 or 28 or 29)31. In Situ Hybridization, Fluorescence/32. fluorescence in situ hybridization.tw.33. urovysion.tw.34. or/31–3335. 6 and 3436. nuclear proteins/37. (nuclear matrix protein 22 or nmp22).tw,rn.38. or/36–3739. 6 and 3840. urine/cy41. urine cytology/use emez42. cytodiagnosis/use mesz43. cancer cytodiagnosis/use emez44. cell count/45. immunocyt$.tw.46. or/40–4547. 6 and 4648. 26 or 30 or 35 or 39 or 4749. exp “costs and cost analysis”/50. economics/51. exp economics,hospital/52. exp economics,medical/53. economics,pharmaceutical/54. exp budgets/55. exp models, economic/155

Appendix 115656. exp decision theory/57. ec.fs. use mesz58. monte carlo method/59. markov chains/60. exp health status indicators/61. cost$.ti.62. (cost$adj2 (effective$or utilit$or benefit$orminimis$)).ab.63. economic$model$.tw.64. (economics$or pharmacoeconomic$orpharmo-economic$).ti.65. (price$or pricing$).tw.66. (financial or finance or finances or financed).tw.67. (value adj2 (money or monetary)).tw.68. markov$.tw.69. monte carlo.tw.70. (decision$adj2 (tree? or analy$or model$)).tw.71. (standard adj1 gamble).tw.72. trade off.tw.73. or/49–7274. 48 and 7375. remove duplicates from 74Science Citation Index (1970 to 1 April 2008)Web of KnowledgeURL: http://wok.mimas.ac.uk/#1 TS=(bladder SAME (cancer* or neoplasm* orcarci*))#2 TS=(hematuria OR haematuria)#3 #1 or #2#4 TS=((bladder SAME resect*) or cystectomy orturbt)#5 #3 or #4#6 TS=(cystoscop* AND (photo* dynamic* ORphotodynamic* OR fluorescence*))#7 #5 AND #6#8 TS=(hypericin or hexvix orhexaminolevulin*or hexyl aminolevulin* or5-ala or 5-aminolevulin*)#9 #5 and #8#10 #7 or #9#11 TS=(marker* SAME (tumor or tumouror biological or molecular or histolog* orbiochem* or genetic* or urine or disease))#12 #3 and #11#13 45,591 TS=(immunocyt* or ucyt)#14 35,989 TS=cytolog*#15 221 TS=(nmp22 or nuclear matrix protein22)#16 33 TS=urovysion#17 13,601 TS=(fluorescence SAMEhybridization)#18 #13 or #14 or #15 or #16 or #17#19 #3 and #18#20 #10 or #12 or #19#21 TS=economic*#22 TS=cost*#23 TS=(price* OR pricing*)#24 TS=(financial or finance*)#25 TS=(decision* SAME (tree* OR analy* ormodel*))#26 TS=markov*#27 TS=monte carlo#28 #21 or #22 or #23 or #24 or #25 or #26 or#27#29 #20 and #28NHS Economic Evaluation Database(March 2008)NHS Centre for Reviews and DisseminationURL:http://nhscrd.york.ac.uk/welcome.htm1 MeSH Bladder Neoplasms EXPLODE 1 2 3 4 49# 2 MeSH Hematuria EXPLODE 1 2 15# 4 nmp22 OR immunocyt OR ucyt OR urovysionOR fish 93# 5 marker* or biomarker* 419# 7 #1 or #2 63# 8 #5 and #7 11# 9 photo AND dynamic OR photodynamic 83# 10 #7 and #9 2# 12 fluorescence OR hexvix OR hexyl ORhypericin OR 5-ala 34# 13 #1 or #2 or #4 or #8 or #10 or #12 171Health Management InformationConsortium (1979 to March 2008)Ovid Multifile SearchURL: http://gateway.ovid.com/athens1. bladder cancer/2. haematuria/3. 1 or 24. (photo$dynamic$or photodynamic orfluorescence).tw. (17)5. (hypericin or hexvix or hexyl$or 5-ala$oraminolevulonate).tw.6. (marker$or biomarker$).tw.7. (nmp22 or immunocyt$or ucyt$or urovysion orfish).tw. (8. cytology/9. or/4–810. 3 and 9CEA Registry (March 2008)Centre for the Evaluation of Value and Risk inHealthURL: https://research.tufts-nemc.org/cear/default.aspxbladder or hemauria or haematuria

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Quality of life and cost data formodelMEDLINE (1966 to March Week 32008), EMBASE (1980 to 2008 Week 13),Medline In-Process (1 April 2008)Ovid Multifile SearchURL: http://gateway.ovid.com/athens1. urinary bladder neoplasms/di, pc2. exp bladder cancer/di, dm3. *hematuria/4. (hematuria or haematuria).ti.5. (bladder adj1 (cancer$or neoplasm$or carci$)).ti.6. *cystoscopy/7. or/1–68. exp “costs and cost analysis”/9. economics/10. exp economics,hospital/11. exp economics,medical/12. economics,pharmaceutical/13. exp budgets/14. exp models, economic/15. exp decision theory/16. ec.fs. use mesz17. monte carlo method/18. markov chains/19. exp health status indicators/20. cost$.ti.21. (cost$adj2 (effective$or utilit$or benefit$orminimis$)).ab.22. economic$model$.tw.23. (economics$or pharmacoeconomic$orpharmo-economic$).ti24. (price$or pricing$).tw.25. (financial or finance or finances or financed).tw.26. (value adj2 (money or monetary)).tw.27. markov$.tw.28. monte carlo.tw.29. (decision$adj2 (tree? or analy$or model$)).tw.30. (standard adj1 gamble).tw.31. trade off.tw.32. or/8–3133. 7 and 3234. quality of life/35. quality adjusted life year/36. “Value of Life”/use mesz37. health status indicators/use mesz38. health status/use emez39. sickness impact profile/use mesz40. disability evaluation/use mesz41. disability/use emez42. activities of daily living/use mesz43. exp daily life activity/use emez44. cost utility analysis/use emez45. rating scale/46. questionnaires/47. (quality adj1 life).tw.48. quality adjusted life.tw.49. disability adjusted life.tw.50. (qaly? or qald? or qale? or qtime? or daly?).tw.51. (euroqol or euro qol or eq5d or eq 5d).tw.52. (hql or hqol or h qol or hrqol or hr qol).tw.53. (hye or hyes).tw.54. health$year$equivalent$.tw.55. (hui or hui1 or hui2 or hui3).tw.56. (health adj3 (utilit$or disutili$)).tw.57. (health adj3 (state or status)).tw.58. (sf36 or sf 36 or short form 36 or shortform36).tw.59. (sf6 or sf 6 or short form 6 or shortform 6).tw.60. (sf12 or sf 12 or short form 12 or shortform12).tw.61. (sf16 or sf 16 or short form 16 or shortform16).tw.62. (sf20 or sf 20 or short form 20 or shortform20).tw.63. willingness to pay.tw.64. standard gamble.tw.65. or/34–6466. 7 and 6567. 33 or 6668. (case report or editorial or letter).pt.69. case report/70. 67 not (68 or 69)71. limit 70 to english language72. remove duplicates from 71IDEAS (March 2008)RePeCURL: http://ideas.repec.org/Bladder or hematuria or haematuriaWebsites consultedCancer Research UK –URL: www.cancerresearchuk.org/European Association of Urology –URL: www.uroweb.org/European Organisation for Research andTreatment of Cancer (EORTC) –URL: www.eortc.be/Hexvix, GE Healthcare Medical Diagnostics –URL: www.hexvix.com/cont.shtml157© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Appendix 1National Cancer Institute, US National Institutes ofHealth –URL: www.cancer.gov/National Comprehensive Cancer Network –URL: www.nccn.org/default.aspNational Insitute of Diabetes and Digestive andKidney Diseases (NIDDK) –URL: http://www2.niddk.nih.gov/Research/ScientificAreas/Urology/NHS National Institute for Health and ClinicalExcellence –URL: www.nice.org.uk/Scottish Intercollegiate Guidelines Network, NHSQuality Improvement Scotland –URL: www.sign.ac.uk/158

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Appendix 2PDD quality assessmentchecklist (QUADAS tool)Study id:Assessor initials:Date assessed:Item Yes No Unclear1 Was the spectrum of patients representative of the patients who will receivethe test in practice?2 Is the reference standard likely to correctly classify the target condition?3 Is the time period between the reference standard and index test shortenough to be reasonably sure that the target condition did not changebetween the two tests?4 Did the whole sample or a random selection of the sample receiveverification using a reference standard of diagnosis?5 Did patients receive the same reference standard regardless of the index testresult?6 Was the reference standard independent of the index test (i.e. the index testdid not form part of the reference standard)?7 Were the index test results interpreted without knowledge of the results ofthe reference standard?8 Were the reference standard results interpreted without knowledge of theresults of the index test?9 Were the same clinical data available when test results were interpreted aswould be available when the test is used in practice?10 Were uninterpretable/intermediate test results reported?11 Were withdrawals from the study explained?12 Did the study provide a clear definition of what was considered to be a‘positive’ result?13 Were data on observer variation reported and within an acceptable range?159© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Appendix 3PDD quality assessment checklist (RCTs)Study id:Assessor initials:Date assessed:Criteria Yes No Unclear1 Was the assignment to the treatment groups really random? (Adequateapproaches to sequence generation: computer-generated random tables,random number tables; inadequate approaches to sequence generation: use ofalternation, case record numbers, birth dates or week days)2 Was the treatment allocation concealed? [Adequate approaches toconcealment of randomisation: centralised or pharmacy-controlledrandomisation, serially numbered identical containers, on-site computer-basedsystem with a randomisation sequence that is not readable until allocation,other approaches with robust methods to prevent foreknowledge of theallocation sequence to clinicians and patients; inadequate approaches toconcealment of randomisation: use of alternation, case record numbers, birthdates or week days, open random numbers lists, serially numbered envelopes(even sealed opaque envelopes can be subject to manipulation)]3 Were the groups similar at baseline in terms of prognostic factors?4 Were the eligibility criteria specified?5 Was the intervention (and comparison) clearly defined?6 Were the groups treated in the same way apart from the interventionreceived?7 Was follow-up long enough to detect important effects on outcomes ofinterest?8 Was the outcome assessor blinded to the treatment allocation?9 Was the care provider blinded?10 Were the patients blinded?11 Were the point estimates and measures of variability presented for theprimary outcome measures?12 Was the withdrawal/dropout rate likely to cause bias?13 Did the analyses include an intention to treat analysis?14 Was the operation undertaken by somebody experienced in performing theprocedure?161© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Appendix 4Photodynamic diagnosis (PDD) included studiesDiagnostic accuracyCheng 2000Cheng CW, Lau WK, Tan PH, Olivo M. Cystoscopicdiagnosis of bladder cancer by intravesical instillation of5-aminolevulinic acid induced porphyrin fluorescence– the Singapore experience. Ann Acad Med Singapore2000;29:153–8.Colombo 2007Colombo R, Naspro R, Bellinzoni P, Fabbri F, GuazzoniG, Scattoni V, et al. Photodynamic diagnosis for follow-upof carcinoma in situ of the bladder. Ther Clin Risk Manage2007;3:1003–7.De Dominicis 2001De Dominicis C, Liberti M, Perugia G, De Nunzio C,Sciobica F, Zuccala A, et al. Role of 5-aminolevulinic acidin the diagnosis and treatment of superficial bladdercancer: improvement in diagnostic sensitivity. Urology2001;57:1059–62.D’Hallewin 2000D’Hallewin MA, De Witte PA, Waelkens E, Merlevede W,Baert L. Fluorescence detection of flat bladder carcinomain situ after intravesical instillation of hypericin. J Urol2000;164:349–51.Ehsan 2001Ehsan A, Sommer F, Haupt G, Engelmann U.Significance of fluorescence cystoscopy for diagnosis ofsuperficial bladder cancer after intravesical instillation ofdelta aminolevulinic acid. Urol Int 2001;67:298–304.Filbeck 1999Primary referenceFilbeck T, Roessler W, Knuechel R, Straub M, KielHJ, Wieland WF. Clinical results of the transurethralresection and evaluation of superficial bladdercarcinomas by means of fluorescence diagnosis afterintravesical instillation of 5-aminolevulinic acid. JEndourol 1999;13:117–21.Secondary referenceFilbeck T, Roessler W, Knuechel R, Straub M, Kiel HJ,Wieland WF. 5-aminolevulinic acid-induced fluorescenceendoscopy applied at secondary transurethral resectionafter conventional resection of primary superficialbladder tumors. Urology 1999;53:77–81.Fradet 2007Fradet Y, Grossman HB, Gomella L, Lerner S, CooksonM, Albala D, et al. A comparison of hexaminolevulinatefluorescence cystoscopy and white light cystoscopyfor the detection of carcinoma in situ in patients withbladder cancer: a phase III, multicenter study. J Urol2007;178:68–73.Frimberger 2001Frimberger D, Zaak D, Stepp H, Knuchel R,Baumgartner R, Schneede P, et al. Autofluorescenceimaging to optimize 5-ALA-induced fluorescenceendoscopy of bladder carcinoma. Urology 2001;58:372–5.Grimbergen 2003Grimbergen MC, van Swol CF, Jonges TG, Boon TA, vanMoorselaar RJ. Reduced specificity of 5-ALA inducedfluorescence in photodynamic diagnosis of transitionalcell carcinoma after previous intravesical therapy. EurUrol 2003;44:51–6.Hendricksen 2006Hendricksen K, Moonen PM, der Heijden AG, Witjes JA.False-positive lesions detected by fluorescence cystoscopy:any association with p53 and p16 expression? World JUrol 2006;24:597–601.Hungerhuber 2007Primary referenceHungerhuber E, Stepp H, Kriegmair M, Stief C,Hofstetter A, Hartmann A, et al. Seven years’ experiencewith 5-aminolevulinic acid in detection of transitionalcell carcinoma of the bladder. Urology 2007;69:260–4.Secondary referencesZaak D, Kriegmair M, Stepp H, Stepp H, BaumgartnerR, Oberneder R, et al. Endoscopic detection oftransitional cell carcinoma with 5-aminolevulinic acid:results of 1012 fluorescence endoscopies. Urology2001;57:690–4.Zaak D, Hungerhuber E, Schneede P, Stepp H,Frimberger D, Corvin S, et al. Role of 5-aminolevulinicacid in the detection of urothelial premalignant lesions.Cancer 2002;95:1234–8.Jeon 2001Jeon SS, Kang I, Hong JH, Choi HY, Chai SE. Diagnosticefficacy of fluorescence cystoscopy for detection ofurothelial neoplasms. J Endourol 2001;15:753–9.163© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Appendix 4Jichlinski 1997Primary referenceJichlinski P, Forrer M, Mizeret J, Glanzmann T,Braichotte D, Wagnieres G, et al. Clinical evaluation of amethod for detecting superficial surgical transitional cellcarcinoma of the bladder by light-induced fluorescenceof protoporphyrin IX following the topical applicationof 5-aminolevulinic acid: preliminary results. Lasers SurgMed 1997;20:402–8.Secondary referenceJichlinski P, Wagnieres G, Forrer M, Mizeret J, GuillouL, Oswald M, et al. Clinical assessment of fluorescencecytoscopy during transurethral bladder resection insuperficial bladder cancer. Urol Res 1997;25(Suppl.1):S3–6.Jichlinski 2003Jichlinski P, Guillou L, Karlsen SJ, Malmstrom PU,Jocham D, Brennhovd B, et al. Hexyl aminolevulinatefluorescence cystoscopy: new diagnostic tool forphotodiagnosis of superficial bladder cancer – amulticenter study. J Urol 2003;170:226–9.Jocham 2005Jocham D, Witjes F, Wagner S, Zeylemaker B, vanMoorselaar J, Grimm MO, et al. Improved detection andtreatment of bladder cancer using hexaminolevulinateimaging: a prospective, phase III multicenter study. JUrol 2005;174:862–6.Koenig 1999Koenig F, McGovern FJ, Larne R, Enquist H,Schomacker KT, Deutsch TF. Diagnosis of bladdercarcinoma using protoporphyrin IX fluorescenceinduced by 5-aminolaevulinic acid. BJU Int 1999;83:129–35.Kriegmair 1996Primary referenceKriegmair M, Baumgartner R, Knuchel R, Stepp H,Hofstadter F, Hofstetter A. Detection of early bladdercancer by 5-aminolevulinic acid induced porphyrinfluorescence. J Urol 1996;155:105–9.Secondary referencesKriegmair M, Baumgartner R, Knuechel R, Steinbach P,Ehsan A, Lumper W, et al. Fluorescence photodetectionof neoplastic urothelial lesions following intravesicalinstillation of 5-aminolevulinic acid. Urology1994;44:836–41.Kriegmair M, Stepp H, Steinbach P, Lumper W, EhsanA, Stepp HG, et al. Fluorescence cystoscopy followingintravesical instillation of 5-aminolevulinic acid: a newprocedure with high sensitivity for detection of hardlyvisible urothelial neoplasias. Urol Int 1995;55:190–6.Kriegmair 1999Primary referenceKriegmair M, Zaak D, Stepp H, Stepp H, BaumgartnerR, Knuechel R, et al. Transurethral resectionand surveillance of bladder cancer supported by5-aminolevulinic acid-induced fluorescence endoscopy.Eur Urol 1999;36:386–92.Secondary referencesSchneeweiss S, Kriegmair M, Stepp H. Is everythingall right if nothing seems wrong? A simple method ofassessing the diagnostic value of endoscopic procedureswhen a gold standard is absent. J Urol 1999;161:1116–9.Schneeweiss S. Sensitivity analysis of the diagnosticvalue of endoscopies in cross-sectional studies in theabsence of a gold standard. Int J Technol Assess Health Care2000;16:834–41.Landry 2003Landry JL, Gelet A, Bouvier R, Dubernard JM, MartinX, Colombel M. Detection of bladder dysplasia using5-aminolaevulinic acid-induced porphyrin fluorescence.BJU Int 2003;91:623–6.Riedl 1999Riedl CR, Plas E, Pfluger H. Fluorescence detection ofbladder tumors with 5-amino-levulinic acid. J Endourol1999;13:755–9.Sim 2005Sim HG, Lau WK, Olivo M, Tan PH, Cheng CW. Isphotodynamic diagnosis using hypericin better thanwhite-light cystoscopy for detecting superficial bladdercarcinoma? BJU Int 2005;95:1215–18.Song 2007Song X, Ye Z, Zhou S, Yang W, Zhang X, Liu J, et al.The application of 5-aminolevulinic acid-inducedfluorescence for cystoscopic diagnosis and treatmentof bladder carcinoma. Photodiagnosis Photodyn Ther2007;4:39–43.Szygula 2004Primary referenceSzygula M, Wojciechowski B, Adamek M, PietrusaA, Kawczyk-Krupka A, Cebula W, et al. Fluorescentdiagnosis of urinary bladder cancer – a comparison oftwo diagnostic modalities. Photodiagnosis Photodyn Ther2004;1:23–6.Secondary referenceSzygula M, Wojciechowski B, Adamek M, Kawczyk-Krupka A, Cebula W, Zieleznik W, et al. Photodynamic vsautofluorescent diagnosis of urinary bladder using XillixLIFE system. Physica Medica 2004;20(Suppl. 1):55–7.164

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Tritschler 2007Tritschler S, Scharf S, Karl A, Tilki D, Knuechel R,Hartmann A, et al. Validation of the diagnostic value ofNMP22 BladderChek test as a marker for bladder cancerby photodynamic diagnosis. Eur Urol 2007;51:403–7.Witjes 2005Witjes JA, Moonen PM, van der Heijden AG.Comparison of hexaminolevulinate based flexible andrigid fluorescence cystoscopy with rigid white lightcystoscopy in bladder cancer: results of a prospectivephase II study. Eur Urol 2005;47:319–22.Zaak 2002Zaak D, Stepp H, Baumgartner R, Schneede P, WaidelichR, Frimberger D, et al. Ultraviolet-excited (308 nm)autofluorescence for bladder cancer detection. Urology2002;60:1029–33.Zumbraegel 2003Zumbraegel A, Bichler KH, Krause FS, Feil G, Nelde HJ.The photodynamic diagnosis (PDD) for early detectionof carcinoma and dysplasia of the bladder. Adv Exp MedBiol 2003;539:61–6.EffectivenessBabjujk 2005Babjuk M, Soukup V, Petrik R, Jirsa M, Dvoracek J.5-aminolaevulinic acid-induced fluorescence cystoscopyduring transurethral resection reduces the risk ofrecurrence in stage Ta/T1 bladder cancer. BJU Int2005;96:798–802.Daniltchenko 2005Primary referenceDaniltchenko DI, Riedl CR, Sachs MD, Koenig F,Daha KL, Pflueger H, et al. Long-term benefit of5-aminolevulinic acid fluorescence assisted transurethralresection of superficial bladder cancer: 5-year results of aprospective randomized study. J Urol 2005;174:2129–33.Secondary referenceRiedl CR, Daniltchenko D, Koenig F, Simak R,Loening SA, Pflueger H. Fluorescence endoscopy with5-aminolevulinic acid reduces early recurrence rate insuperficial bladder cancer. J Urol 2001;165:1121–3.Denzinger 2007Primary referenceDenzinger S, Burger M, Walter B, Knuechel R, RoesslerW, Wieland WF, et al. Clinically relevant reduction inrisk of recurrence of superficial bladder cancer using5-aminolevulinic acid-induced fluorescence diagnosis:8-year results of prospective randomized study. Urology2007;69:675–9.Secondary referencesBurger M, Zaak D, Stief CG, Filbeck T, Wieland WF,Roessler W, et al. Photodynamic diagnostics andnoninvasive bladder cancer: is it cost-effective in longtermapplication? A Germany-based cost analysis. EurUrol 2007;52:142–7.Denzinger S, Wieland WF, Otto W, Filbeck T, KnuechelR, Burger M. Does photodynamic transurethral resectionof bladder tumour improve the outcome of initial T1high-grade bladder cancer? A long-term follow-up of arandomized study. BJU Int 2008;101:566–9.Filbeck T, Pichlmeier U, Knuechel R, Wieland WF,Roessler W. Clinically relevant improvement ofrecurrence-free survival with 5-aminolevulinic acidinduced fluorescence diagnosis in patients withsuperficial bladder tumors. J Urol 2002;168:67–71.Kriegmair 2002Kriegmair M, Zaak D, Rothenberger KH, Rassweiler J,Jocham D, Eisenberger F, et al. Transurethral resectionfor bladder cancer using 5-aminolevulinic acid inducedfluorescence endoscopy versus white light endoscopy. JUrol 2002;168:475–8.165© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Appendix 5Photodynamic diagnosis excluded studiesRequired outcomes notreported (n = 12)Chin WW, Ramaswamy B, Thong PSP, Heng PWS, GanYY, Olivo M, et al. Preclinical and pilot clinical cancerstudies using fluorescence-guided photodynamic therapywith chlorin e6-polyvinylpyrrolidone and hypericin.Singapore Gen Hosp Proc 2007;16:118–26.D’Hallewin MA, Vanherzeele H, Baert L. Fluorescencedetection of flat transitional cell carcinoma afterintravesical instillation of aminolevulinic acid. Am J ClinOncol 1998;21:223–5.D’Hallewin MA, Kamuhabwa AR, Roskams T, De WittePA, Baert L. Hypericin-based fluorescence diagnosis ofbladder carcinoma. BJU Int 2002;9:760–3.Filbeck T, Pichlmeier U, Knuechel R, Wieland WF,Roessler W. Do patients profit from 5-aminolevulinicacid-induced fluorescence diagnosis in transurethralresection of bladder carcinoma? Urology 2002;60:1025–8.Grossman HB, Gomella L, Fradet Y, Morales A, Presti J,Ritenour C, et al. A phase III, multicenter comparisonof hexaminolevulinate fluorescence cystoscopy andwhite light cystoscopy for the detection of superficialpapillary lesions in patients with bladder cancer. J Urol2007;178:62–7.Junker K, Kania K, Fiedler W, Hartmann A, Schubert J,Werner W. Molecular genetic evaluation of fluorescencediagnosis in bladder cancer. Int J Oncol 2002;20:647–53.Kriegmair M, Zaak D, Knuechel R, Baumgartner R,Hofstetter A. 5-Aminolevulinic acid-induced fluorescenceendoscopy for the detection of lower urinary tracttumors. Urol Int 1999;63:27–31.Kriegmair M, Zaak D, Knuechel R, Baumgartner R,Hofstetter A. Photodynamic cystoscopy for detection ofbladder tumors. Semin Laparosc Surg 1999;6:100–3.Loidl W, Schmidbauer J, Susani M, Marberger M.Flexible cystoscopy assisted by hexaminolevulinateinduced fluorescence: a new approach for bladder cancerdetection and surveillance? Eur Urol 2005;47:323–6.Petrik R, Jirsa M, Dvorak E, Skoda V, Stadnik B.Fluorescence cystoscopy in the diagnostics andtreatment of bladder tumors. Biomed Tech (Berl)1998;43(Suppl.):74–5.Schmidbauer J, Witjes F, Schmeller N, Donat R, SusaniM, Marberger M, et al. Improved detection of urothelialcarcinoma in situ with hexaminolevulinate fluorescencecystoscopy. J Urol 2004;171:135–8.van der Meijden APM. Does hexaminolevulinate imagingimprove the detection and treatment of bladder cancer?Nature Clin Pract Urol 2006;3:22–3.Required study design not met(n = 10)Batlle A, Peng Q. Preface: special issue of photodynamictherapy and photodetection with porphyrin precursorsfor the Journal of Environmental Pathology, Toxicology, andOncology. J Env Pathol Toxicol Oncol 2007;26:ix–xii.Chatterton K, Ray E, O’Brien TS. Fluorescence diagnosisof bladder cancer. Br J Nurs 2006;15:595–7.Collaud S, Jichlinski P, Marti A, Aymon D, GurnyR, Lange N. An open pharmacokinetic study ofhexylaminolevulinate-induced photodiagnosis afterintravesical administration. Drugs R D 2006;7:173–86.Grossman H. Re: Long-term benefit of 5-aminolevulinicacid fluorescence-assisted transurethral resection ofsuperficial bladder cancer: 5-year results of a prospectiverandomised study. Eur Urol 2006;50:861–2.Jichlinski P. Hexyl aminolevulinate in the detection ofbladder cancer. Drugs 2006;66:579–80.Koenig F, McGovern FJ. Fluorescence detection ofbladder carcinoma. Urology 1997;50:778–9.Witjes JA, Jichlinski P, Zaak D, Stief CG. Hexylaminolevulinate in the detection of bladder cancer. Drugs2006;66:579–80.Zaak D, Frimberger D, Stepp H, Wagner S, BaumgartnerR, Schneede P, et al. Quantification of 5-aminolevulinicacid induced fluorescence improves the specificity ofbladder cancer detection. J Urol 2001;166:1665–8.Zaak D, Stief CG. Hexyl aminolevulinate in the detectionof bladder cancer. Drugs 2006;66(4).Zlotta A. Fluorescence cystoscopy: is flexible scope aseffective as rigid? Eur Urol 2005;47:318.167© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Appendix 5Required reference standard notmet (n = 2)Olivo M, Lau W, Manivasager V, Hoon TP, ChristopherC. Fluorescence confocal microscopy and image analysisof bladder cancer using 5-aminolevulinic acid. Int J Oncol2003;22:523–8.Olivo M, Lau W, Manivasager V, Tan PH, Soo KC, ChengC. Macro-microscopic fluorescence of human bladdercancer using hypericin fluorescence cystoscopy and laserconfocal microscopy. Int J Oncol 2003;23:983–90.Comparator test not white lightcystoscopy (n = 1)Lipinski M, Jeromin L. Comparison of the bladdertumour antigen test with photodynamic diagnosisin patients with pathologically confirmed recurrentsuperficial urinary bladder tumours. BJU Int2002;89:757–9.168

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Appendix 6Characteristics of the PDD diagnostic studiesStudy a Participants Tests Outcomes summaryCheng 2000 50Time period: Jan 1997 toDec 1998Country: SingaporeEnrolled: 41; analysed: 41No previous history of BC:NS; history of BC: NSAge (years): mean 66.8, range42 to 89Sex: M 24; F 17Index test: PDDAgent: 5-ALAComparator: WLC‘Random’ biopsies of normalappearingareas: yes for PDDUnit of analysis: biopsy(n = 175)Sensitivity: PDD 89%, WLC66%Specificity: PDD 65%, WLC84%Colombo 2007 51Time period: Feb 2004 toMar 2006Country: ItalyEnrolled: 49; analysed: 49No previous history of BC: 0;history of BC: 49Age (years): mean 70, SD 12Sex: NSNotes: All patients weresuffering from CIS alone atinclusion and undergoingBCG therapyIndex test: PDDAgent: 5-ALA, HALComparator: WLC‘Random’ biopsies of normalappearingareas: yes (NSwhether PDD or WLC orboth)Unit of analysis: patient(n = 49)Sensitivity: PDD 100%, WLC0%Specificity: PDD 71%, WLC97%De Dominicis 2001 53Time period: May 1997 toNSCountry: ItalyEnrolled: 49; analysed: 49No previous history of BC:17; history of BC: 32Age (years): mean 60, range31 to 77Sex: M 42; F 7Index test: PDDAgent: 5-ALAComparator: WLC‘Random’ biopsies of normalappearingareas: yes for bothPDD and WLCUnit of analysis: biopsy(n = 179)Sensitivity: PDD 87%, WLC17%Specificity: PDD 63%, WLC88%D’Hallewin 2000 52Time period: NSCountry: BelgiumEnrolled: 40; analysed: 40No previous history of BC:NS; history of BC: NSAge (years): NSSex: NSIndex test: PDDAgent: hypericinComparator: none‘Random’ biopsies of normalappearingareas: yes for PDDUnit of analysis: biopsy (CIS)(n = 281)Sensitivity: PDD 93%Specificity: PDD 99%Ehsan 2001 54Time period: NSCountry: GermanyEnrolled: 30; analysed: 30No previous history of BC:NS; history of BC: NSAge (years): mean NS, range55 to 85Sex: M 19; F 11Index test: PDDAgent: 5-ALAComparator: WLC‘Random’ biopsies of normalappearingareas: yes for PDDand WLCUnit of analysis: biopsy(n = 151)Sensitivity: PDD 59%, WLC60%Specificity: PDD 98%, WLC58%Filbeck 1999 56Time period: NSCountry: GermanyEnrolled: 123; analysed: 120No previous history of BC:NS; history of BC: NSAge (years): mean 64.5, range28 to 86Sex: NSNotes: 60 of the patientswere having a secondaryresection 6 weeks afterprimary tumour resectionIndex test: PDDAgent: 5-ALAComparator: WLC‘Random’ biopsies of normalappearingareas: no (exceptin cases of a resection inareas of a primary tumour)Unit of analysis: biopsy(n = 347)Sensitivity: PDD 96%Specificity: PDD 35%169© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Appendix 6Study a Participants Tests Outcomes summary[Filbeck 1999 55 ]Time period: Jan 1997 toOct 1997Country: GermanyEnrolled: 50; analysed: 50No previous history of BC:NS; history of BC: NSAge (years): mean 63.4, range32 to 88Sex: M 36, F 14Notes: Patients hadundergone conventional TURof primary tumour 6 weeksearlierIndex test: PDDAgent: 5-ALAComparator: WLC‘Random’ biopsies of normalappearingareas: no for PDDand WLCUnit of analysis: biopsy(n = 347)Sensitivity: WLC 69%Specificity: WLC 66%Unit of analysis: biopsy(n = 130)Sensitivity: PDD 78%Specificity: PDD 33%Unit of analysis: biopsy(n = 18)Sensitivity: WLC 64%Specificity: NSFradet 2007 57Time period: NSCountry: USA, CanadaEnrolled: 311; analysed: 196(1 NS?)No previous history of BC:62; history of BC: 133Age (years): mean 67, SD 11Sex: M 148, F 48Notes: 49 patients receivedprevious chemotherapy and77 received previous BCGtreatmentIndex test: PDDAgent: HALComparator: WLC‘Random’ biopsies of normalappearingareas: yes for PDDand WLCUnit of analysis: patient(n = 196)Sensitivity: PDD 87%, WLC83%Specificity: PDD 82%, WLC72%Unit of analysis: biopsy(n = NS, CIS 113)Sensitivity: PDD 92%, WLC68%Specificity: NSFrimberger 2001 58Time period: NSCountry: GermanyEnrolled: 25; analysed: 25No previous history of BC: 0;history of BC: 25Age (years): NSSex: NSIndex test: PDDAgent: 5-ALAComparator: WLC‘Random’ biopsies of normalappearingareas: yes for PDDand WLCUnit of analysis: biopsy(n = 19)Sensitivity: PDD 95%Specificity: PDD 67%NS for WLCGrimbergen 2003 59Time period: Nov 1998 toJun 2002Country: NetherlandsEnrolled: 160; analysed: 160No previous history of BC:87?; history of BC: 73?Age (years): mean 67, range30 to 91Sex: NSNotes: 73 patients receivedprevious BCG, mitomycin Cor epirubicin treatmentIndex test: PDDAgent: 5-ALAComparator: WLC‘Random’ biopsies of normalappearingareas: yes for PDDand WLCUnit of analysis: biopsy(n = 917)Sensitivity: PDD 97%, WLC69%Specificity: PDD 49%, WLC78%Hendricksen 2006 60Time period: Oct 2001 toApr 2002Country: NetherlandsEnrolled: 50; analysed: 50No previous history of BC:23; history of BC: 27Age (years): mean 67, range35 to 86Sex: M 40, F 10Notes: This study takesthe patient data from theRadbound University MedicalCentre, Nijmesen thatcontributed to Jocham 2005and Schmidbauer 2004Index test: PDDAgent: HALComparator: WLC‘Random’ biopsies of normalappearingareas: yes (NSwhether PDD or WLC orboth)Unit of analysis: biopsy (PDDn = 217, WLC n = 123)Sensitivity: PDD 94%, WLC88%Specificity: PDD 58%, WLC86%170

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Study a Participants Tests Outcomes summaryHungerhuber 2007 61Time period: Feb 1995 toFeb 2002Country: GermanyEnrolled: 875; analysed: 875No previous history of BC:327; history of BC: 548Age (years): mean 65.3, range16 to 99Sex: M 671, F 204Notes: Patients with ahistory of recurrent diseasehad undergone multipleTURs (mean 3.6, range 1 to22)Index test: PDDAgent: 5-ALAComparator: WLC‘Random’ biopsies of normalappearingareas: no for PDDand WLCUnit of analysis: biopsy(n = 4630)Sensitivity: PDD 92%, WLC76%Specificity: PDD 56%, WLC86%[Zaak 2002 83 ]Time period: Jan 1995 toDec 2000Country: Germany, AustriaEnrolled: 713; analysed: 713No previous history of BC:270; history of BC: 443Age (years): NSSex: NSNotes: Patients previouslytreated for BC had a historyof undergoing multiple TURs(mean 3.5, range 1 to 20)Index test: PDDAgent: 5-ALAComparator: WLC‘Random’ biopsies of normalappearingareas: no for PDDand WLCUnit of analysis: biopsy (PDDn = 3834, WLC NS)Sensitivity: PDD 98%, WLC47%Specificity: PDD 21%, WLCNS[Zaak 2001 82 ]Time period: 1995 to 1999Country: GermanyEnrolled: 605; analysed: 605No previous history of BC:212; history of BC: 393Age (years): mean 65.6, range16 to 99Sex: M 472, F 133Notes: Patients previouslytreated for BC had a historyof undergoing multiple TURs(mean 3.5, range 1 to 20)Index test: PDDAgent: 5-ALAComparator: WLC‘Random’ biopsies of normalappearingareas: no for PDDand WLCUnit of analysis: biopsy (PDDn = 1012, WLC n = 552)Sensitivity: PDD 86%, WLC66%Specificity: PDD 23%, WLCNSJeon 2001 62Time period: Dec 1997 toAug 1999Country: South KoreaEnrolled: 62; analysed: 62No previous history of BC:36; history of BC: 26Age (years): mean 61.9, range32 to 80Sex: M 57, F 5Notes: Of the patientswith a history of BC, fivehad nephrourterectomyperformed with a bladdercuff resection for upperurinary tract carcinoma andsix had BCGIndex test: PDDAgent: 5-ALAComparator: WLC‘Random’ biopsies of normalappearingareas: no for PDDand WLCUnit of analysis: biopsy(n = 274)Sensitivity: PDD 98%, WLC61Specificity: PDD 41%, WLC92%Jichlinski 1997 63Time period: Feb 1994 to NSCountry: SwitzerlandEnrolled: 34; analysed: 34No previous history of BC:13; history of BC: 21Age (years): mean 67.9, range44 to 84Sex: M 21, F 13Index test: PDDAgent: 5-ALAComparator: WLC‘Random’ biopsies of normalappearingareas: yes for PDDonlyUnit of analysis: biopsy(n = 215)Sensitivity: PDD 89%, WLC46%?Specificity: PDD 57% WLC57%?171© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Appendix 6Study a Participants Tests Outcomes summary[Jichlinski 1997 64 ]Time period: Jan 1995 to NSCountry: SwitzerlandEnrolled: 31; analysed: 31No previous history of BC:11; history of BC: 22Age (years): mean 66.1, range44 to 84Sex: M 23, F 8Notes: Topical chemotherapyor immunotherapy with BCGwas added to the previoussurgical treatments in 19patientsIndex test: PDDAgent: 5-ALAComparator: WLC‘Random’ biopsies of normalappearingareas: biopsies ofapparently normal mucosaunder WLCUnit of analysis: biopsy(n = 132)Sensitivity: PDD 83%Specificity: PDD 81%Jichlinski 2003 65Time period: Dec 2000 toApr 2001Country: Switzerland,Norway, Sweden, GermanyEnrolled: 52; analysed: 52No previous history of BC:18; history of BC: 34Age (years): mean 72, SD 12Sex: M 38, F 14Index test: PDDAgent: HALComparator: WLC‘Random’ biopsies of normalappearingareas: yes for WLCUnit of analysis: patient(n = 52)Sensitivity: PDD 96%, WLC73%Specificity: PDD 43%, WLC43%Unit of analysis: biopsy (PDDn = 421, WLC n = 414)Sensitivity: PDD 76%, WLC80%Specificity: PDD 46%, WLC93%Jocham 2005 66Time period: NSCountry: Germany,NetherlandsEnrolled: 162; analysed: 146No previous history of BC:73; history of BC: 73Age (years): mean 67, range33 to 91Sex: M 107, F 39Notes: 18% received previousBCG immunotherapy and18% received previousintravesical chemotherapyIndex test: PDDAgent: HALComparator: WLC‘Random’ biopsies of normalappearingareas: no for PDDand WLCUnit of analysis: patient(n = 146)Sensitivity: PDD 53%, WLC33%Specificity: PDD 81%, WLC74%Koenig 1999 67Time period: NSCountry: Germany, USAEnrolled: 55; analysed: 49No previous history of BC:NS; history of BC: NSAge (years): mean 66, range31 to 87Sex: M 44, F 11Index test: PDDAgent: 5-ALAComparator: WLC‘Random’ biopsies of normalappearingareas: yes (NSwhether PDD or WLC orboth)Unit of analysis: biopsy (PDDn = 130, WLC n = 67)Sensitivity: PDD 87%, WLC84%Specificity: PDD 59%, WLCNS172

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Study a Participants Tests Outcomes summaryKriegmair 1996 70Time period: NSCountry: GermanyEnrolled: 106; analysed: 106No previous history of BC:29; history of BC: 77Age (years): mean 68, range41 to 85Sex: M 80, F 24Index test: PDDAgent: 5-ALAComparator: WLC‘Random’ biopsies of normalappearingareas: yes for PDDUnit of analysis: biopsy(n = 433)Sensitivity: PDD 98%, WLC73%Specificity: PDD 64%, WLC69%Unit of analysis: patient(n = 308 – all patients)Sensitivity: PDD 93%, WLC70%Specificity: PDD 53%, WLC75%Unit of analysis: patient(n = 165 – history of BCG orchemotherapy)Sensitivity: PDD 96%, WLC62%Specificity: PDD 72%, WLC71%[Kriegmair 1994 68 ]Time period: NSCountry: GermanyEnrolled: 68; analysed: 68No previous history of BC: 6;history of BC: 62Age (years): mean 66.2, range43 to 83Sex: M 51, F 17Notes: 47 patients receivedprevious intravesicalchemotherapy or BCGIndex test: PDDAgent: 5-ALAComparator: none‘Random’ biopsies of normalappearingareas: yes for PDDUnit of analysis: biopsy(n = 285)Sensitivity: PDD 100%Specificity: PDD 76%[Kriegmair 1995 69 ]Time period: NSCountry: GermanyEnrolled: 90; analysed: 90No previous history of BC:26; history of BC: 64Age (years): mean 65, range41 to 85Sex: NSNotes: 64 patients withhistory of BC had receivedprevious intravesical therapywith BCG or cytostaticsIndex test: PDDAgent: 5-ALAComparator:‘Random’ biopsies of normalappearingareas: yes for PDDUnit of analysis: biopsy(n = 294)Sensitivity: PDD 98%Specificity: PDD 71%Kriegmair 1999 71Time period: NSCountry: GermanyEnrolled: 208; analysed: 208No previous history of BC:72; history of BC: 136Age (years): mean 64.8, range16 to 89Sex: M 170, F 38Notes: Patients previouslytreated for BC had a historyof multiple TURS (mean3.5, range 1 to 20) andintravesical instillation withBCG (n = 50) or mitomycinC (n = 49)Index test: PDDAgent: 5-ALAComparator: WLC‘Random’ biopsies of normalappearingareas: no for PDDand WLCUnit of analysis: biopsy (PDDn = 328, WLC n = 163)Sensitivity: PDD 98%, WLC47%Specificity: PDD 41%, WLCNS173© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Appendix 6Study a Participants Tests Outcomes summary[Schneeweiss 1999 74 ]Time period: Jan 1995 to Aug1996Country: Germany[Schneeweiss 2000 75 ]As aboveEnrolled: 208; analysed: 208No previous history of BC:72; history of BC: 136Age (years): mean 64.8, SD12.4, range 16 to 89Sex: M 170, F 38Index test: PDDAgent: 5-ALAComparator: WLC‘Random’ biopsies of normalappearingareas: no for PDDand WLCUnit of analysis: biopsy(n = 328)Sensitivity: PDD 98%, WLC47%Specificity: PDD 41%, WLCNSLandry 2003 72Time period: NSCountry: FranceEnrolled: 50; analysed: 50No previous history of BC:50; history of BC: 0Age (years): NSSex: NSIndex test: PDDAgent: 5-ALAComparator:‘Random’ biopsies of normalappearingareas: yes for WLCUnit of analysis: patient(n = 50)Sensitivity: PDD 64%, WLCNSSpecificity: PDD 67%, WLCNSRiedl 1999 73Time period: NSCountry: AustriaEnrolled: 52; analysed: 52No previous history of BC:NS; history of BC: NSAge (years): range 44 to 79Sex: NSIndex test: PDDAgent: 5-ALAComparator: WLC‘Random’ biopsies of normalappearingareas: yes (NSwhether PDD or WLC orboth)Unit of analysis: patient(n = 52)Sensitivity: PDD 100%, WLC76%Specificity: PDD 67%, WLC100%Unit of analysis: biopsy(n = 123)Sensitivity: PDD 95%, WLC76%Specificity: PDD 43%, WLCNSSim 2005 76Time period: Jan 2001 to Oct2004Country: SingaporeEnrolled: 41; analysed: 41No previous history of BC:NS; history of BC: NSAge (years): mean 66.1, SD9.1, range 46 to 81Sex: M 34, F 7Index test: PDDAgent: hypericinComparator: WLC‘Random’ biopsies of normalappearingareas: no for PDDand WLCUnit of analysis: biopsy(n = 179)Sensitivity: PDD 82%, WLC62%Specificity: PDD 91%, WLC98%Song 2007 77Time period: Mar 2002 toOct 2005Country: ChinaEnrolled: 51; analysed: 51No previous history of BC:47; history of BC: 4Age (years): mean 52Sex: M 32, F 19Notes: All patients had typicalwhole range anodynia grosshaematuriaIndex test: PDDAgent: 5-ALAComparator: WLC‘Random’ biopsies of normalappearingareas: no for PDDand WLCUnit of analysis: patient (PDDn = 51, WLC n = 40)Sensitivity: PDD 100%, WLC53%Specificity: PDD 36%, WLCNSSzygula 2004 78Time period: NSCountry: PolandEnrolled: 52 (PDD group);analysed: 52No previous history of BC:52; history of BC: 0Age (years): NSSex: NSNotes: All patients receivedTURBT 3 months beforeinvestigative procedure. Allpatients received WLCIndex test: PDDAgent: 5-ALAComparator: LIF‘Random’ biopsies of normalappearingareas: noNotes: unclear whethercomparing PDD with LIF orPDD + WLC with LIF; noWLC only comparisonUnit of analysis: patient(n = 52)Sensitivity: PDD 91%, WLCNSSpecificity: PDD 67%, WLCNS174[Szygula 2004 79 ]As above

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Study a Participants Tests Outcomes summaryTritschler 2007 80Time period: Sep 2004 toApr 2005Country: GermanyEnrolled: 100; analysed: 100No previous history of BC:30; history of BC: 70Age (years): mean 67.9Sex: M 71, F 29Index test: PDDAgent: 5-ALA/HALComparator: WLC‘Random’ biopsies of normalappearingareas: no for PDDand WLCUnit of analysis: patient(n = 100)Sensitivity: PDD 93%, WLC88%Specificity: PDD 57%, WLC55%Witjes 2005 81Time period: Jan 2004 to Mar2004Country: NetherlandsEnrolled: 20; analysed: 20No previous history of BC:10; history of BC: 10Age (years): mean 71, range49 to 89Sex: M 17, F 3Notes: Seven patientsreceived previous intravesicalchemotherapy or BCG forsuperficial papillary tumoursIndex test: PDDAgent: HALComparator: WLC‘Random’ biopsies of normalappearingareas: no for PDDand WLCUnit of analysis: patient(n = 20)Sensitivity: PDD 90%, WLC79%Specificity: PDD 100%, WLC100%Unit of analysis: biopsy(n = 28)Sensitivity: PDD 85%, WLC74%Specificity: PDD 100%, WLC100%Zaak 2002 84Time period: NSCountry: GermanyEnrolled: 43; analysed: 43No previous history of BC: 0;history of BC: 43Age (years): mean 70, range49 to 89Sex: M 31, F 12Index test: PDDAgent: 5-ALAComparator: excimer laserinducedautofluorescence; noWLC comparison‘Random’ biopsies of normalappearingareas: yes for PDDUnit of analysis: biopsy(n = 114)Sensitivity: PDD 90%Specificity: PDD 61%Zumbraegel 2003 85Time period: Jan 1997 to Jul1999Country: GermanyEnrolled: 108; analysed: 152No previous history of BC:NS; history of BC: NSAge (years): NSSex: NSIndex test: PDDAgent: 5-ALAComparator: WLC‘Random’ biopsies of normalappearingareas: no for PDDor WLCUnit of analysis: biopsy(n = 408)Sensitivity: PDD 94%, WLC80%Specificity: PDD 32%, WLC46%BC, bladder cancer; BCG, bacillus Calmette–Guerin; LIF, laser-induced fluorescence; NS, not stated.a Studies in square brackets, e.g. [Jichlinski 1997], are secondary reports.175© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Appendix 7Quality assessment results forthe individual PDD studies177© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Appendix 7178Diagnostic studiesStudy Q1 Q2 Q3 Q4 Q5 Q6 Q7 Q8 Q9 Q10 Q11 Q12 Q13Cheng 2000 50 + + + + + – + ? ? + + + –Colombo 2007 51 – + + + + – + + ? + + – –De Dominicis 2001 53 + + + + + – + ? ? + + + –D’Hallewin 2000 52 + + + + + – + + ? + + + –Ehsan 2001 54 + + + + + – + ? + + + + –Filbeck 1999 56 + + + + – – + ? ? + + + –Fradet 2007 57 + + + + + – + + ? + + – –Frimberger 2001 58 + + + + + – + ? ? + + + –Grimbergen 2003 59 + + + + + – + + ? + + + –Hendricksen 2006 60 + + + + + – + ? ? + + + –Hungerhuber 2007 61 + + + + – – + + ? + + + –Jeon 2001 62 + + + + – – + + ? + + + –Jichlinski 1997 63 + + + + + – + + ? + + + –Jichlinski 2003 65 + + + + + – + + ? + + + –Jocham 2005 66 + + + + – – + + ? + + + –Koenig 1999 67 + + + + + – + ? ? – – + –Kriegmair 1996 70 + + + + + – + ? ? + + + –Kriegmair 1999 71 + + + + – – + ? ? + + + –Landry 2003 72 + + + + + – + ? ? + + + –Riedl 1999 73 + + + + ? – + ? ? + + + –Sim 2005 76 + + + + – – + + ? + + + –Song 2007 77 + + + + – – + ? ? + + + –Szygula 2004 78 + + + + – – + ? ? + + + –Tritschler 2007 80 + + + + – – + + ? + + – –Witjes 2005 81 + + + + – – + ? ? + + – –Zaak 2002 84 + + + + + – + ? ? + + + –Zumbraegel 2003 85 + + + + – – + ? ? + + – –+, yes to the question; –, no to the question; ?, unclear.

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4RCTs reporting recurrence/progressionStudy Q1 Q2 Q3 Q4 Q5 Q6 Q7 Q8 Q9 Q10 Q11 Q12 Q13 Q14Babjuk 2005 86 ? ? + + + + + ? ? ? – – – ?Daniltchenko 2005 88 ? ? + + + ? + ? ? ? + – – ?Denzinger 2007 89 ? ? + + + + + ? ? ? + + – ?Kriegmair 2002 92 ? ? + + + ? + ? ? ? – + + ?+, yes to the question; –, no to the question; ?, unclear.179© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Appendix 8Studies of PDD versus WLC included inpooled estimates for patient- and biopsy-levelanalysis and also those reporting stage/grade181© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Appendix 8Patient Biopsy pTa pTaG1 pTaG1–2 pTaG2 pTaG2–3 pTaG3 pTa–T1 G1–2Cheng 2000 50 Colombo 2007 51De Dominicis 2001 53 Ehsan 2001 54 Filbeck 1999 56 Fradet 2007 57Grimbergen 2003 59Hendricksen 2006 60 Hungerhuber 2007 61 Jeon 2001 62Jichlinski 1997 63Jichlinski 2003 65 Jocham 2005 66 PKoenig 1999 67 Kriegmair 1996 70 Riedl 1999 73 Sim 2005 76Tritschler 2007 80 P PWitjes 2005 81 P,B P,BZumbraegel 2003 85P, patient-level analysis; B, biopsy-level analysis.182

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4pT1 pT1G1 pT1G1–2 pT1G2 pT1G3 > pT1 CIS G3 pT2G2 pT2G3 ≥ pT2 ≥ pT2G3 pT4G3P P,B P,BP P P P P P PP,B P,B P,B183© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Appendix 9PDD and WLC test performance for detectingbladder cancer, results table with 2 × 2 data185© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Appendix 9186Spec(%) LR+ LR–Sens(%)Numberanalysed TP FP FN TNStudy a,b,c Test Unit of analysisPDD Biopsy 175 57 39 7 72 89 65 2.5 0.2(5-ALA)Flat lesions 136 20 38 7 71 74 65 2.1 0.4CIS lesions 7 5 2 71pTa–T1 lesions 36 36 0 100≥ pT2 lesions 5 5 0 100WLC Biopsy 175 42 18 22 93 66 84 4.0 0.4Flat lesions 136 5 16 22 93 23 85 1.5 0.9CIS lesions 7 1 6 14pTa–T1 lesions 36 30 6 83≥ pT2 lesions 5 5 0 100Cheng 2000 50No. of patients 41, ofwhom primary NS,recurrent NSPDD ‘random’ biopsiesPatients with CIS 49 18 9 0 22 100 71 3.4 0.0PDD(5-ALA,HAL)Colombo 2007 51WLC Patients with CIS 49 0 1 18 30 0 97 0.0 0.0No. of patients 49, ofwhom primary 0, recurrent49 (all being followed upfor CIS)‘Random’ biopsies (NSwhether PDD or WLC orboth)PDD Biopsy 179 45 47 7 80 87 63 2.3 0.2(5-ALA)Dysplasia 19 16 3 84CIS 13 10 3 77SNA 2 1 1 50pTaG2–3 10 10 0 100pT1G3 0pT1G2–3 8 8 0 100WLC Biopsy 179 9 15 43 112 17 88 1.5 0.9Dysplasia 19 6 13 32CIS 13 3 10 23SNA 2 0 2 0pTaG2–3 10 0 10 0pT1G3 8 0 8 0pT1G2–3 0De Dominicis 2001 53No. of patients 49, ofwhom primary 17,recurrent 32PDD and WLC ‘random’biopsies

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Spec(%) LR+ LR–Sens(%)Numberanalysed TP FP FN TNStudy a,b,c Test Unit of analysisBiopsyCIS 281 132 2 10 137 93 99 93.0 0.1PDD(hypericin)D’Hallewin 2000 52No. of patients 40, ofwhom primary NS,recurrent NSPDD ‘random’ biopsiesPDD Biopsy 151 59 32 1 59 98 65 2.8 0.0(5-ALA)CIS 2 2 0 100pTa 20 20 0 100pT1 28 28 0 100WLC Biopsy 151 36 38 24 53 60 58 1.4 0.7CIS 2 0 2 0pTa 20 18 2 90pT1 28 14 14 50Ehsan 2001 54No. of patients 30, ofwhom primary NS,recurrent NSPDD and WLC ‘random’biopsies347 119 145 5 78 96 35 1.5 0.1Biopsy (primary/recurrent/secondary tumour resection)PDD(5-ALA)Filbeck 1999 56pTaG1–2 77 74 3 96pT1G1–2 10 9 1 90pT1G3 11 11 0 100CIS 7 7 0 100> pT1 6 6 0 100Dysplasia 13 12? 1? 92?Biopsy (primary/recurrenttumour resection)No. of patients 120,of whom primary NS,recurrent NS (60 of thepatients were havinga secondary resection6 weeks after primarytumour resection)No ‘random’ biopsies(except in cases of aresection in areas ofprimary tumour)pTaG1–2 64 64 0 100pT1G1–2 8 8 0 100pT1G3 10 10 0 100CIS 5 5 0 100> pT1 6 6 0 100Dysplasia 6 6 0 100187© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Appendix 9188Spec(%) LR+ LR–Sens(%)Numberanalysed TP FP FN TNStudy a,b,c Test Unit of analysisBiopsy (secondary tumourresection)pTaG1–2 13 10 3 77pT1G1–2 2 1 1 50pT1G3 1 1 0 100CIS 2 2 0 100Dysplasia 7 6 1 86WLC Biopsy (primary/recurrent/ 347 85 75 39 148 69 66 2.0 0.5secondary tumour resection)pTaG1–2 77 57 20 74pT1G1–2 10 8 2 80pT1G3 11 8 3 73CIS 7 4 3 57> pT1 6 5 1 83Dysplasia 13 3 10 30Biopsy (primary/recurrenttumour resection)pTaG1–2 64 54 10 84pT1G1–2 8 7 1 88pT1G3 10 8 2 80CIS 5 4 1 80>pT1 6 5 1 83Dysplasia 6 2 4 33Biopsy (secondary tumourresection)pTaG1–2 13 3 10 23pT1G1–2 2 1 1 50pT1G3 1 0 1 0CIS 2 0 2 0Dysplasia 7 1 6 14

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Spec(%) LR+ LR–Sens(%)Numberanalysed TP FP FN TNStudy a,b,c Test Unit of analysisBiopsy 130 14 75 4 37 78 33 1.2 0.7PDD(5-ALA)WLC Biopsy 18 7? 11? 64?[Filbeck 1999 55 ] (secondaryTUR)No. of patients 50, allof whom were having asecondary TUR of theformer resection area 6weeks after conventionalTURNo ‘random’ biopsiesPDD (HAL)PatientCIS 196 50 25 8 113 87 82 4.8 0.2BiopsyCIS 113 104 223 9 92WLC PatientCIS 196 48 39 10 99 83 72 3.0 0.2BiopsyCIS 113 77 137 36 68Fradet 2007 57No. of patients 196,of whom primary 62,recurrent 133 (missing, 1)PDD and WLC ‘random’biopsiesBiopsy 43 18 8 1 16 95 67 2.9 0.1CIS 6 5 1 83pTa 10 10 0 100pT2 3 3 0 100PDD(5-ALA)Frimberger 2001 58No. of patients 25, all ofwhom were recurrentPDD and WLC ‘random’biopsiesBiopsy 917 378 270 12 257 97 49 1.9 0.1PDD(5-ALA)Grimbergen 2003 59WLC Biopsy 917 270 115 120 412 69 78 3.2 0.4No. of patients 160,of whom primary 87?,recurrent 73?PDD and WLC ‘random’biopsies189© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Appendix 9190Spec(%) LR+ LR–Sens(%)Numberanalysed TP FP FN TNStudy a,b,c Test Unit of analysisBiopsy 217 88 52 6 71 94 58 2.2 0.1CIS 10 7 3 70pTaG1 2 2 0 100pTaG2 39 37 2 95pTaG3 6 6 0 100pT1G2 13 12 1 92pT1G3 8 8 0 100pT2G2 2 2 0 100pT2G3 8 8 0 100pT4G3 3 3 0 100PDD(HAL)Hendricksen 2006 60No. of patients 50, ofwhom primary 23,recurrent 27‘Random’ biopsies (unclearwhether PDD or WLC orboth)WLC Biopsy 123 83 17 11 106 88 86 6.4 0.1CIS 10 3 7 30pTaG1 2 2 0 100pTaG2 39 39 0 100pTaG3 6 6 0 100pT1G2 13 12 1 92pT1G3 8 7 1 88pT2G2 2 2 0 100pT2G3 8 8 0 100pT4G3 3 3 0 100

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Spec(%) LR+ LR–Sens(%)Numberanalysed TP FP FN TNStudy a,b,c Test Unit of analysisPDD Biopsy 4630 1484 1339 129 1678 92 56 2.1 0.1(5-ALA)pTxG1 47 42 5 89pTxG2 28 26 2 93pTxG3 18 15 3 83pTaG1 495 466 29 94pTaG2 205 196 9 96pTaG3 45 42 3 93pT1G1 12 11 1 92pT1G2 53 52 1 98pT1G3 149 144 5 97≥ T2G3 101 90 11 89CIS 274 254 20 93Dysplasia grade 2 186 146 40 78WLC Biopsy 4630 1231 430 382 2587 76 86 5.4 0.3pTxG1 47 40 7 85pTxG2 28 23 5 82pTxG3 18 15 3 83pTaG1 495 408 87 82pTaG2 205 175 30 85pTaG3 45 34 11 76pT1G1 12 8 4 67pT1G2 53 38 15 72pT1G3 149 119 30 80≥ T2G3 101 87 14 86CIS 274 155 119 57Dysplasia grade 2 186 129 57 69Hungerhuber 2007 61No. of patients 875, ofwhom primary 327,recurrent 548No ‘random’ biopsies191© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Appendix 9192Spec(%) LR+ LR–Sens(%)Numberanalysed TP FP FN TNStudy a,b,c Test Unit of analysisPDD Lesion (same as biopsy) 3834 1222 2049 28 535 98 21 1.2 0.1(5-ALA)CIS 159 145 14 91WLC Lesion (same as biopsy)CIS 159 75 84 47[Zaak 2002 83 ]No. of patients 713, ofwhom primary 270,recurrent 443No ‘random’ biopsiesPDD Endoscopy 1012 477 352 75 108 86 23 1.1 0.6(5-ALA)Dysplasia grade 2 52 30 22 58CIS 88 84 4 95pTxGx 1 1 0 100pTxG1 13 11 2 85pTxG2 5 5 0 100pTxG3 6 6 0 100pTaG1 178 146 32 82pTaG2 80 69 11 86pTaG3 11 11 0 100pT1G1 9 8 1 89pT1G2 26 26 0 100pT1G3 46 43 3 93> pT1G2–3 37 37 0 100WLC Endoscopy 552 363 189 66Dysplasia grade 2 52 32 20 62CIS 88 38 50 43pTxGx 1 1 0 100pTxG1 13 12 1 92pTxG2 5 3 2 60pTxG3 6 5 1 69pTaG1 178 118 60 66pTaG2 80 61 19 76pTaG3 11 6 5 55pT1G1 9 6 3 67pT1G2 26 15 11 58pT1G3 46 34 12 74> pT1G2–3 37 32 5 86[Zaak 2001 82 ]No. of patients 605, ofwhom primary 212,recurrent 393No ‘random’ biopsies

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Spec(%) LR+ LR–Sens(%)Numberanalysed TP FP FN TNStudy a,b,c Test Unit of analysisPDD (5-ALA)BiopsyCIS or more 274 140 77 3 54 98 41 1.7 0.0Dysplasia or more 274 145 72 3 54 98 43 1.7 0.0Dysplasia 5 5 0 100CIS 20 20 0 100Ta 64 63 1 98T1 42 40 2 95≥ T2 17 17 0 100WLC BiopsyCIS or more 274 87 11 56 120 61 92 7.6 0.4Dysplasia or more 274 87 11 61 115 59 91 6.6 0.5Dysplasia 5 0 5 0CIS 20 1 19 5Ta 64 48 16 75T1 42 25 17 60≥ T2 17 13 4 76Jeon 2001 62No. of patients 62, ofwhom primary 36,recurrent 26No ‘random’ biopsiesBiopsy 215 97 46 12 60 89 57 2.1 0.2PDD(5-ALA)Jichlinski 1997 63WLC Biopsy 215 50? 46?? 59? 60?? 46? 57? 1.1 0.9No. of patients 34, ofwhom primary 13,recurrent 21PDD ‘random’ biopsies132 34 17 7 74 83 81 4.4 0.2Biopsies of apparently normalmucosa under WLCPDD(5-ALA)[Jichlinski 1997 64 ]No. of patients 31, ofwhom primary 11,recurrent 22 (data asreported)WLC ‘random’ biopsies193© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Appendix 9194Spec(%) LR+ LR–Sens(%)Numberanalysed TP FP FN TNStudy a,b,c Test Unit of analysisPDD Patient 52 43 4 2 3 96 43 1.7 0.1(HAL)CIS 13 12 1 92Biopsy 421 108 57 35 221 76 79 3.7 0.3CIS 57 31? 26? 54?WLC Patient 52 33 4 12 3 73 43 1.3 0.6CIS 13 3 10 30Biopsy 414 65 19 75 255 46 93 6.7 0.6CIS 57 3 54 5Jichlinski 2003 65No. of patients 52, ofwhom primary 18,recurrent 34WLC ‘random’ biopsiesPDD Patient 146 61 6 54 25 53 81 2.7 0.6(HAL)CIS 29 12 17 41pTa 66 13 53 20pT1 (pT1a + pT1b) 16 1 15 6pT2–4 22 3 19 14WLC Patient 146 38 8 77 23 33 74 1.3 0.9CIS 29 2 27 7pTa 66 5 61 8pT1 (pT1a + pT1b) 16 1 15 6pT2–4 22 1 21 5Jocham 2005 66No. of patients 146,of whom primary 73,recurrent 73No ‘random’ biopsiesBiopsy 130 58 26 9 37 87 59 2.1 0.2CIS 6 5 1 83pTaG1 8 7 1 88pTaG2 25 22 3 88pTaG3 3 3 0 100pT1G2 2 2 0 100pT1G3 5 5 0 100pT2G3 6 5 1 83PDD(5-ALA)Koenig 1999 67No. of patients 55, ofwhom primary NS,recurrent NS‘Random’ biopsies (unclearwhether PDD or WLC orboth)

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Spec(%) LR+ LR–Sens(%)Numberanalysed TP FP FN TNStudy a,b,c Test Unit of analysisWLC Biopsy 67 56 11 84CIS 6 4 2 67pTaG1 8 6 2 75pTaG2 25 24 1 96pTaG3 0pT1G2 0pT1G3 0pT2G3 6 5 1 83Biopsy 433 126 110 2 195 98 64 2.7 0.0Dysplasia grade 1 + CIS 329 23 110 1 195 96 64 2.7 0.1Dysplasia grade 2 + CIS 221 13 12 0 196 100 94 17.2 0.0CIS 329 6 127 0 196 100 61 2.5 0.0pTa–T1 399 93 110 1 195 99 64 2.7 0.0Patients with dysplasia or 308 28 83 2 195 93 70 3.1 0.1TCC from normal bladderwall or bladder wall withnon-specific inflammation –all patientsPatients with dysplasia or TCC from normal bladderwall or bladder wall withnon-specific inflammation– patients with a historyof BCG instillation orchemotherapy165 25 39 1 100 96 72 3.4 0.1PDD(5-ALA)Kriegmair 1996 70No. of patients 106,of whom primary 29,recurrent 77PDD ‘random’ biopsies195© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Appendix 9196Spec(%) LR+ LR–Sens(%)Numberanalysed TP FP FN TNStudy a,b,c Test Unit of analysisWLC Biopsy 433 93 96 35 209 73 69 2.3 0.4Dysplasia grade 1 + CIS 329 10 96 14 209 42 69 1.3 0.9Dysplasia grade 2 + CIS 329 7 99 6 217 54 69 1.7 0.7CIS 329 4 102 2 221 67 68 2.1` 0.5pTa–T1 399 74 96 20 209 79 69 2.5 0.3Patients with dysplasia or 308 16 69 14 209 53 75 2.1 0.6TCC from normal bladderwall or bladder wall withnon-specific inflammation –all patientsPatients with dysplasia or TCC from normal bladderwall or bladder wall withnon-specific inflammation– patients with a historyof BCG instillation orchemotherapy165 16 41 10 98 62 71 2.1 0.5Biopsy 285 93 46 0 146 100 76 4.2 N/CPDD(5-ALA)[Kriegmair 1994 68 ]No. of patients 68, ofwhom primary 6, recurrent62PDD ‘random’ biopsiesBiopsy 294 43 73 1 177 98 71 3.4 0.0PDD(5-ALA)[Kriegmair 1995 69 ]No. of patients 90, ofwhom primary 26,recurrent 64PDD ‘random’ biopsiesBiopsy 328 159 97 4 68 98 41 1.7 0.0PDD(5-ALA)Kriegmair 1999 71,74,75WLC Biopsy 163 77 86 47No. of patients 208,of whom primary 72,recurrent 136No ‘random’ biopsiesPatient 50 9 12 5 24 64 67 1.9 0.5PDD(5-ALA)Landry 2003 72No. of patients 50, ofwhom primary 50,recurrent 0WLC ‘random’ biopsies

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Spec(%) LR+ LR–Sens(%)Numberanalysed TP FP FN TNStudy a,b,c Test Unit of analysisPDD Patient 52 34 12 0 6 100 33 1.5 N/C(5-ALA)Biopsy 123 50? 40? 3? 30? 95 43 1.7 0.1WLC Patient 52 26 0 8 18 76 100 N/C 0.2Biopsy 70 53? 17? 76Riedl 1999 73No. of patients 52, ofwhom primary NS,recurrent NS‘Random’ biopsies (unclearwhether PDD or WLC orboth)Biopsy 179 61 10 13 95 82 90 8.7 0.2PDD(hypericin)Sim 2005 76WLC Biopsy 179 46 2 28 103 62 98 32.7 0.4No. of patients 41, ofwhom primary NS,recurrent NSNo ‘random’ biopsiesPDD Patient 51 40 7 0 4 100 36 1.6 N/C(5-ALA)Biopsy 103 68 21 2 12 97 36 1.5 0.1WLC Patient 40 21 19 53Song 2007 77No. of patients 51, ofwhom primary 47,recurrent 4No ‘random’ biopsiesPatient 52 20 10 2 20 91 67 2.7 0.1PDD(5-ALA)Szygula 2004 78,79No. of patients 52, ofwhom primary 52?,recurrent 0?No ‘random’ biopsiesPatient 100 37 26 3 34 93 57 2.1 0.1pTa 22? 21? 1? 95CIS 9? 7? 2? 78pT1 4? 4? 0? 100?≥ pT2 3? 3? 0? 100?G1–2 22? 21? 1? 95?G3 18? 16? 2? 89?PDD(5-ALA/HAL)Tritschler 2007 80No. of patients 100,of whom primary 30,recurrent 70No ‘random’ biopsies197© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Appendix 9198Spec(%) LR+ LR–Sens(%)Numberanalysed TP FP FN TNStudy a,b,c Test Unit of analysisWLC Patient 100 35 27 5 33 88 55 1.9 0.2pTa 22? 21? 1? 95?CIS 9? 5? 4? 56?pT1 4? 4? 0? 100?≥ pT2 3? 3? 0? 100?G1–2 22? 21? 1? 95?G3 18? 14? 4? 78?Patient 20 17 0 2 1 89 100 N/C 0.1pTaG2 8 7 1 88pTaG3 1 1 0 100CIS 3 2 1 67pT1G3 1 1 0 100pT2G3 6 6 0 100Lesion (same as biopsy) 28 23 0 4 1 85 100 N/C 0.1pTaG2 11 10 1 91pTaG3 1 1 0 100CIS 8 5 3 63pT1G3 1 1 0 100pT2G3 6 6 0 100PDD(HAL)Witjes 2005 81No. of patients 20, ofwhom primary 10,recurrent 10No ‘random’ biopsiesWLC Patient 20 15 0 4 1 79 100 N/C 0.2pTaG2 8 8 0 100pTaG3 1 1 0 100CIS 3 1 2 33pT1G3 1 1 0 100pT2G3 6 4 2 67Lesion 28 20 0 7 1 74 100 N/C 0.3pTaG2 1 10 1 91pTaG3 1 1 0 100CIS 8 4 4 50pT1G3 1 1 0 100pT2G3 6 4 2 67

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Spec(%) LR+ LR–Sens(%)Numberanalysed TP FP FN TNStudy a,b,c Test Unit of analysisBiopsy 114 19 36 2 57 90 61 2.3 0.2PDD (5-ALA)Zaak 2002 84No. of patients 43, ofwhom primary 0, recurrent43PDD ‘random’ biopsiesPDD (5- Biopsy 408 125 186 8 89 94 32 1.4 0.1ALA)CIS 14 12 2 86Dysplasia 48 43 5 90WLC Biopsy 408 107 148 26 127 80 46 1.5 0.4CIS 14 9 5 64Dysplasia 48 15 33 31Zumbraegel 2003 85No. of patients 108,of whom primary NS,recurrent NSNo ‘random’ biopsiesFN, false negative; FP, false positive; LR+, positive likelihood ratio; LR–, negative likelihood ratio; NC, not calculable; NS, not stated; SNA, single nuclear atypia; TN, true negative; TP, truepositive.a Studies in square brackets, e.g. [Jichlinski 1997], are secondary reports.b Blank cells: no data reported.c No ‘random’ biopsies: study either stated that no random biopsies were carried out or did not state whether random biopsies were carried out. PDD ‘random’ biopsies: biopsies ofnormal-appearing areas on PDD. WLC ‘random’ biopsies: biopsies of normal-appearing areas on WLC.199© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Appendix 10Biomarker/cytology included studiesAbbate 1998Abbate I, D’Introno A, Cardo G, Marano A, Addabbo L,Musci MD, et al. Comparison of nuclear matrix protein22 and bladder tumor antigen in urine of patients withbladder cancer. Anticancer Res 1998;18:3803–5.Bastacky 1999Bastacky S, Ibrahim S, Wilczynski SP, Murphy WM. Theaccuracy of urinary cytology in daily practice. Cancer1999;87:118–28.Bhuiyan 2003Bhuiyan J, Akhter J, O’Kane DJ. Performancecharacteristics of multiple urinary tumor markersand sample collection techniques in the detection oftransitional cell carcinoma of the bladder. Clin Chim Acta2003;331:69–77.Boman 2002Boman H, Hedelin H, Holmang S. Four bladder tumormarkers have a disappointingly low sensitivity for smallsize and low grade recurrence. J Urol 2002;167:80–3.Casella 2000Primary referenceCasella R, Huber P, Blochlinger A, Stoffel F, Dalquen P,Gasser TC, et al. Urinary level of nuclear matrix protein22 in the diagnosis of bladder cancer: experiencewith 130 patients with biopsy confirmed tumor. J Urol2000;164:1926–8.Secondary referencesShariat SF, Casella R, Monoski MA, Sulser T, GasserTC, Lerner SP. The addition of urinary urokinase-typeplasminogen activator to urinary nuclear matrix protein22 and cytology improves the detection of bladdercancer. J Urol 2003;170:2244–7.Shariat SF, Casella R, Khoddami SM, Hernandez G,Sulser T, Gasser TC, et al. Urine detection of survivin is asensitive marker for the noninvasive diagnosis of bladdercancer. J Urol 2004;171:626–30.Casetta 2000Casetta G, Gontero P, Zitella A, Pelucelli G, FormiconiA, Priolo G, et al. BTA quantitative assay and NMP22testing compared with urine cytology in the detectionof transitional cell carcinoma of the bladder. Urol Int2000;65:100–5.Chahal 2001aChahal R, Gogoi NK, Sundaram SK. Is it necessaryto perform urine cytology in screening patients withhaematuria? Eur Urol 2001;39:283–6.Chahal 2001bChahal R, Darshane A, Browning AJ, Sundaram SK.Evaluation of the clinical value of urinary NMP22 as amarker in the screening and surveillance of transitionalcell carcinoma of the urinary bladder. Eur Urol2001;40:415–20.Chang 2004Chang YH, Wu CH, Lee YL, Huang PH, Kao YL,Shiau MY. Evaluation of nuclear matrix protein-22 asa clinical diagnostic marker for bladder cancer. Urology2004;64:687–92.Daniely 2007Daniely M, Rona R, Kaplan T, Olsfanger S, ElboimL, Freiberger A, et al. Combined morphologic andfluorescence in situ hybridization analysis of voidedurine samples for the detection and follow-up of bladdercancer in patients with benign urine cytology. Cancer2007;111:517–24.Del Nero 1999Del Nero A, Esposito N, Curro A, Biasoni D, MontanariE, Mangiarotti B, et al. Evaluation of urinary levelof NMP22 as a diagnostic marker for stage pTa-pT1bladder cancer: comparison with urinary cytology andBTA test. Eur Urol 1999;35:93–7.Friedrich 2003Primary referenceFriedrich MG, Toma MI, Hellstern A, Pantel K,Weisenberger DJ, Noldus J, et al. Comparison ofmultitarget fluorescence in situ hybridization in urinewith other noninvasive tests for detecting bladder cancer.BJU Int 2003;92:911–4.Secondary referenceFriedrich MG, Hellstern A, Hautmann SH, Graefen M,Conrad S, Huland E, et al. Clinical use of urinary markersfor the detection and prognosis of bladder carcinoma:a comparison of immunocytology with monoclonalantibodies against Lewis X and 486p3/12 with the BTASTAT and NMP22 tests. J Urol 2002;168:470–4.201© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Appendix 10202Garbar 2007Garbar C, Mascaux C, Wespes E. Is urinary tractcytology still useful for diagnosis of bladder carcinomas?A large series of 592 bladder washings using a fivecategoryclassification of different cytological diagnoses.Cytopathology 2007;18:79–83.Giannopoulos 2001Primary referenceGiannopoulos A, Manousakas T, Gounari A,Constantinides C, Choremi-Papadopoulou H,Dimopoulos C. Comparative evaluation of the diagnosticperformance of the BTA stat test, NMP22 and urinarybladder cancer antigen for primary and recurrentbladder tumors [see comment]. J Urol 2001;166:470–5.Secondary referenceGiannopoulos A, Manousakas T, Mitropoulos D, Botsoli-Stergiou E, Constantinides C, Giannopoulou M, et al.Comparative evaluation of the BTAstat test, NMP22, andvoided urine cytology in the detection of primary andrecurrent bladder tumors. Urology 2000;55:871–5.Grossman 2005Grossman HB, Messing E, Soloway M, Tomera K, KatzG, Berger Y, et al. Detection of bladder cancer using apoint-of-care proteomic assay. JAMA 2005;293:810–16.Grossman 2006Grossman HB, Soloway M, Messing E, Katz G, Stein B,Kassabian V, et al. Surveillance for recurrent bladdercancer using a point-of-care proteomic assay. JAMA2006;295:299–305.Guttierez Banos 2001Gutierrez Banos JL, Rebollo Rodrigo MH, Antolin JuarezFM, Martin GB. NMP 22, BTA stat test and cytology inthe diagnosis of bladder cancer: a comparative study.Urol Int 2001;66:185–90.Hakenberg 2000Hakenberg OW, Franke P, Froehner M, Manseck A, WirthMP. The value of conventional urine cytology in thediagnosis of residual tumour after transurethral resectionof bladder carcinomas. Onkologie 2000;23:252–7.Halling 2000Halling KC, King W, Sokolova IA, Meyer RG, BurkhardtHM, Halling AC, et al. A comparison of cytology andfluorescence in situ hybridization for the detection ofurothelial carcinoma. J Urol 2000;164:1768–75.Hughes 1999Hughes JH, Katz RL, Rodriguez-Villanueva J, Kidd L,Dinney C, Grossman HB, et al. Urinary nuclear matrixprotein 22 (NMP22): a diagnostic adjunct to urinecytologic examination for the detection of recurrenttransitional-cell carcinoma of the bladder. DiagnCytopathol 1999;20:285–90.Junker 2006Junker K, Fritsch T, Hartmann A, Schulze W, Schubert J.Multicolor fluorescence in situ hybridization (M-FISH)on cells from urine for the detection of bladder cancer.Cytogenet Genome Res 2006;114:279–83.Karakiewicz 2006Primary referenceKarakiewicz PI, Benayoun S, Zippe C, Ludecke G,Boman H, Sanchez-Carbayo M, et al. Institutionalvariability in the accuracy of urinary cytology forpredicting recurrence of transitional cell carcinoma ofthe bladder. BJU Int 2006;97:997–1001.Secondary referenceHutterer GC, Karakiewicz PI, Zippe C, Ludecke G,Boman H, Sanchez-Carbayo M, et al. Urinary cytologyand nuclear matrix protein 22 in the detection ofbladder cancer recurrence other than transitional cellcarcinoma. BJU Int 2008;101:561–5.Kipp 2008Kipp BR, Halling KC, Campion MB, Wendel AJ,Karnes RJ, Zhang J, et al. Assessing the value of reflexfluorescence in situ hybridization testing in the diagnosisof bladder cancer when routine urine cytologicalexamination is equivocal. J Urol 2008;179:1296–301.Kowalska 2005Kowalska M, Kaminska J, Kotowicz B, Fuksiewicz M,Rysinska A, Demkow T, et al. Evaluation of the urinarynuclear matrix protein (NMP22) as a tumour marker inbladder cancer patients. Nowotwory 2005;55:300–2.Kumar 2006Kumar A, Kumar R, Gupta NP. Comparison of NMP22BladderChek test and urine cytology for the detection ofrecurrent bladder cancer. Jpn J Clin Oncol 2006;36:172–5.Lahme 2001Primary referenceLahme S, Bichler KH, Feil G, Krause S. Comparison ofcytology and nuclear matrix protein 22 for the detectionand follow-up of bladder cancer. Urol Int 2001;66:72–7.Secondary referenceLahme S, Bichler KH, Feil G, Zumbragel A, Gotz T.Comparison of cytology and nuclear matrix protein 22(NMP 22) for the detection and follow-up of bladdercancer.Adv Exp Med Biol 2003;539:111–19.Lee 2001Lee KH. Evaluation of the NMP22 test and comparisonwith voided urine cytology in the detection of bladdercancer. Yonsei Med J 2001;42:14–18.Lodde 2003Lodde M, Mian C, Negri G, Berner L, Maffei N,Lusuardi L, et al. Role of uCyt+ in the detectionand surveillance of urothelial carcinoma. Urology2003;61:243–7.

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Lodde 2006Lodde M, Mian C, Comploj E, Palermo S, Longhi E,Marberger M, et al. uCyt+ test: alternative to cystoscopyfor less-invasive follow-up of patients with low risk ofurothelial carcinoma. Urology 2006;67:950–4.May 2007May M, Hakenberg OW, Gunia S, Pohling P, HelkeC, Lubbe L, et al. Comparative diagnostic value ofurine cytology, UBC-ELISA, and fluorescence in situhybridization for detection of transitional cell carcinomaof urinary bladder in routine clinical practice. Urology2007;70:449–53.Meiers 2007Meiers I, Singh H, Hossain D, Lang K, Liu L, QianJQ, et al. Improved filter method for urine sedimentdetection of urothelial carcinoma by fluorescence in situhybridization. Arch Pathol Lab Med. 2007;131:1574–7.Messing 2005Messing EM, Teot L, Korman H, Underhill E,Barker E, Stork B, et al. Performance of urine test inpatients monitored for recurrence of bladder cancer:a multicenter study in the United States. J Urol2005;174:1238–41.Mian 1999Mian C, Pycha A, Wiener H, Haitel A, Lodde M,Marberger M. Immunocyt: a new tool for detectingtransitional cell cancer of the urinary tract. J Urol1999;161:1486–9.Mian 2000Mian C, Lodde M, Haitel A, Vigl EE, Marberger M,Pycha A. Comparison of the monoclonal UBC-ELISAtest and the NMP22 ELISA test for the detectionof urothelial cell carcinoma of the bladder. Urology2000;55:223–6.Mian2003Mian C, Lodde M, Comploj E, Negri G, Egarter-Vigl E,Lusuardi L, et al. Liquid-based cytology as a tool for theperformance of uCyt+ and Urovysion Multicolour-FISHin the detection of urothelial carcinoma. Cytopathology2003;14:338–42.Mian 2006Mian C, Maier K, Comploj E, Lodde M, Berner L,Lusuardi L, et al. uCyt+/ImmunoCyt in the detectionof recurrent urothelial carcinoma: an update on 1991analyses. Cancer 2006;108:60–5.Miyanaga 1999Miyanaga N, Akaza H, Tsukamoto T, Ishikawa S,Noguchi R, Ohtani M, et al. Urinary nuclear matrixprotein 22 as a new marker for the screening ofurothelial cancer in patients with microscopic hematuria.Int J Urol 1999;6:173–7.Miyanaga 2003Miyanaga N, Akaza H, Tsukamoto S, Shimazui T, OhtaniM, Ishikawa S, et al. Usefulness of urinary NMP22 todetect tumor recurrence of superficial bladder cancerafter transurethral resection. Int J Clin Oncol 2003;8:369–73.Moonen 2007Moonen PM, Merkx GF, Peelen P, Karthaus HF, SmeetsDF, Witjes JA. UroVysion compared with cytology andquantitative cytology in the surveillance of non-muscleinvasivebladder cancer. Eur Urol 2007;51:1275–80.Oge 2001Oge O, Atsu N, Kendi S, Ozen H. Evaluation ofnuclear matrix protein 22 (NMP22) as a tumor markerin the detection of bladder cancer. Int Urol Nephrol2001;32:367–70.Olsson 2001Olsson H, Zackrisson B. ImmunoCyt a useful method inthe follow-up protocol for patients with urinary bladdercarcinoma. Scand J Urol Nephrol 2001;35:280–2.Oosterhuis 2002Oosterhuis JWA, Pauwels RPE, Schapers RFM, Van PeltJ, Smeets W, Newling DWW. Detection of recurrenttransitional cell carcinoma of the bladder with nuclearmatrix protein-22 in a follow-up setting. UroOncology2002;2:137–42.Parekattil 2003Parekattil SJ, Fisher HA, Kogan BA. Neural networkusing combined urine nuclear matrix protein-22,monocyte chemoattractant protein-1 and urinaryintercellular adhesion molecule-1 to detect bladdercancer. J Urol 2003;169:917–20.Piaton 2003Primary referencePiaton E, Daniel L, Verriele V, Dalifard I, ZimmermannU, Renaudin K, et al., French Prospective MS. Improveddetection of urothelial carcinomas with fluorescenceimmunocytochemistry (uCyt+ assay) and urinarycytology: results of a French Prospective MulticenterStudy. Lab Invest 2003;83:845–52.Secondary referencePfister C, Chautard D, Devonec M, Perrin P, ChopinD, Rischmann P, et al. Immunocyt test improves thediagnostic accuracy of urinary cytology: results of aFrench multicenter study. J Urol 2003;169:921–4.Planz 2005Planz B, Jochims E, Deix T, Caspers HP, Jakse G,Boecking A. The role of urinary cytology for detection ofbladder cancer. Eur J Surg Oncol 2005;31:304–8.203© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Appendix 10Ponsky 2001Ponsky LE, Sharma S, Pandrangi L, Kedia S, Nelson D,Agarwal A, et al. Screening and monitoring for bladdercancer: refining the use of NMP22. J Urol 2001;166:75–8.Potter 1999Potter JM, Quigley M, Pengelly AW, Fawcett DP, MalonePR. The role of urine cytology in the assessment of lowerurinary tract symptoms. BJU Int 1999;84:30–1.Poulakis 2001Poulakis V, Witzsch U, De Vries R, Altmannsberger HM,Manyak MJ, Becht E. A comparison of urinary nuclearmatrix protein-22 and bladder tumour antigen testswith voided urinary cytology in detecting and followingbladder cancer: the prognostic value of false-positiveresults. BJU Int 2001;88:692–701.Raitanen 2002Primary referenceRaitanen MP, Aine RA, Kaasinen ES, Liukkonen TJ,Kylmala TM, Huhtala H, et al. Suspicious urine cytology(class III) in patients with bladder cancer: should it beconsidered as negative or positive? Scand J Urol Nephrol2002;36:213–7.Secondary referenceRaitanen M-P, Aine R, Rintala E, Kallio J, Rajala P,Juusela H, et al. Differences between local and reviewurinary cytology in diagnosis of bladder cancer.An interobserver multicenter analysis. Eur Urol2002;41:284–9.Ramakumar 1999Ramakumar S, Bhuiyan J, Besse JA, Roberts SG, WollanPC, Blute ML, et al. Comparison of screening methods inthe detection of bladder cancer. J Urol 1999;161:388–94.Saad 2002Saad A, Hanbury DC, McNicholas TA, Boustead GB,Morgan S, Woodman AC. A study comparing variousnoninvasive methods of detecting bladder cancer inurine. BJU Int 2002;89:369–73.Sanchez-Carbayo 1999Primary referenceSanchez-Carbayo M, Herrero E, Megias J, Mira A, SoriaF. Comparative sensitivity of urinary CYFRA 21–1,urinary bladder cancer antigen, tissue polypeptideantigen, tissue polypeptide antigen and NMP22 to detectbladder cancer. J Urol 1999;162:1951–6.Secondary referenceSanchez-Carbayo M, Herrero E, Megias J, Mira A, SoriaF. Evaluation of nuclear matrix protein 22 as a tumourmarker in the detection of transitional cell carcinoma ofthe bladder. BJU Int 1999;84:706–13.Sanchez-Carbayo 2001aSanchez-Carbayo M, Urrutia M, Gonzalez de BuitragoJM, Navajo JA. Utility of serial urinary tumor markersto individualize intervals between cystoscopies in themonitoring of patients with bladder carcinoma. Cancer2001;92:2820–8.Sanchez-Carbayo 2001bSanchez-Carbayo M, Urrutia M, Silva JM, Romani R, DeBuitrago JM, Navajo JA. Comparative predictive valuesof urinary cytology, urinary bladder cancer antigen,CYFRA 21–1 and NMP22 for evaluating symptomaticpatients at risk for bladder cancer. J Urol 2001;165:1462–7.Sarosdy 2002Sarosdy MF, Schellhammer P, Bokinsky G, Kahn P, ChaoR, Yore L, et al. Clinical evaluation of a multi-targetfluorescent in situ hybridization assay for detection ofbladder cancer. J Urol 2002;168:1950–4.Sarosdy 2006Sarosdy MF, Kahn PR, Ziffer MD, Love WR, Barkin J,Abara EO, et al. Use of a multitarget fluorescence insitu hybridization assay to diagnose bladder cancer inpatients with hematuria. J Urol 2006;176:44–7.Schmitz-Drager 2008Primary referenceSchmitz-Drager BJ, Tirsar LA, Schmitz-Drager C,Dorsam J, Mellan Z, Bismarck E, et al. Immunocytologyin the assessment of patients with asymptomatichematuria. World J Urol 2008;26:31–7.Secondary referenceSchmitz-Drager BJ, Beiche B, Tirsar LA, Schmitz-Drager C, Bismarck E, Ebert T. Immunocytologyin the assessment of patients with asymptomaticmicrohaematuria. Eur Urol 2007;51:1582–8.Serretta 2000Primary referenceSerretta V, Pomara G, Rizzo I, Esposito E. Urinary BTAstat,BTA-trak and NMP22 in surveillance after TUR ofrecurrent superficial transitional cell carcinoma of thebladder. Eur Urol 2000;38:419–25.Secondary referenceSerretta V, Lo PD, Vasile P, Gange E, Esposito E, MenozziI. Urinary NMP22 for the detection of recurrence aftertransurethral resection of transitional cell carcinomaof the bladder: experience on 137 patients. Urology1998;52:793–6.Shariat 2006Shariat SF, Marberger MJ, Lotan Y, Sanchez-Carbayo M,Zippe C, Ludecke G, et al. Variability in the performanceof nuclear matrix protein 22 for the detection of bladdercancer. J Urol 2006;176:919–26.204

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Sharma 1999Sharma S, Zippe CD, Pandrangi L, Nelson D, AgarwalA. Exclusion criteria enhance the specificity andpositive predictive value of NMP22 and BTA stat. J Urol1999;162:53–7.Skacel 2003Skacel M, Fahmy M, Brainard JA, Pettay JD, BiscottiCV, Liou LS, et al. Multitarget fluorescence in situhybridization assay detects transitional cell carcinoma inthe majority of patients with bladder cancer and atypicalor negative urine cytology. J Urol 2003;169:2101–5.Sokolova 2000Sokolova IA, Halling KC, Jenkins RB, Burkhardt HM,Meyer RG, Seelig SA, et al. The development of amultitarget, multicolor fluorescence in situ hybridizationassay for the detection of urothelial carcinoma in urine. JMol Diagn 2000;2:116–23.Sozen 1999Sozen S, Biri H, Sinik Z, Kupeli B, Alkibay T, BozkirliI. Comparison of the nuclear matrix protein 22 withvoided urine cytology and BTA stat test in the diagnosisof transitional cell carcinoma of the bladder. Eur Urol1999;36:225–9.Stampfer 1998Stampfer DS, Carpinito GA, Rodriguez-Villanueva J,Willsey LW, Dinney CP, Grossman HB, et al. Evaluation ofNMP22 in the detection of transitional cell carcinoma ofthe bladder. J Urol 1998;159:394–8.Takeuchi 2004Takeuchi Y, Sawada Y. A clinical study of urinaryNMP22 in urinary epithelial cancer. J Med Soc Toho Univ2004;516:332–8.Talwar 2007Talwar R, Sinha T, Karan SC, Doddamani D, SandhuA, Sethi GS, et al. Voided urinary cytology in bladdercancer: is it time to review the indications? Urology2007;70:267–71.Tetu 2005Tetu B, Tiguert R, Harel F, Fradet Y. ImmunoCyt/uCyt+ improves the sensitivity of urine cytology inpatients followed for urothelial carcinoma. Mod Pathol2005;18:83–9.Tritschler 2007Tritschler S, Scharf S, Karl A, Tilki D, Knuechel R,Hartmann A, et al. Validation of the diagnostic value ofNMP22 BladderChek test as a marker for bladder cancerby photodynamic diagnosis. Eur Urol 2007;51:403–7.Wiener 1998Wiener HG, Mian C, Haitel A, Pycha A, Schatzl G,Marberger M. Can urine bound diagnostic tests replacecystoscopy in the management of bladder cancer? J Urol1998;159:1876–80.Yoder 2007Yoder BJ, Skacel M, Hedgepeth R, Babineau D, UlchakerJC, Liou LS, et al. Reflex UroVysion testing of bladdercancer surveillance patients with equivocal or negativeurine cytology: a prospective study with focus on thenatural history of anticipatory positive findings. Am J ClinPathol 2007;127:295–301.Zippe 1999Primary referenceZippe C, Pandrangi L, Agarwal A. NMP22 is a sensitive,cost-effective test in patients at risk for bladder cancer. JUrol 1999;161:62–5.Secondary referenceZippe C, Pandrangi L, Potts JM, Kursh E, NovickA, Agarwal A. NMP22: a sensitive, cost-effective testin patients at risk for bladder cancer. Anticancer Res1999;19:2621–3.205© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Appendix 11Biomarker/cytology excluded studiesLess than 100 participantsincluded in the analysis (n = 119)Abd El Gawad IA, Moussa HS, Nasr MI, El Gemae EH,Masooud AM, Ibrahim IK, et al. Comparative studyof NMP-22, telomerase, and BTA in the detection ofbladder cancer. J Egypt Natl Cancer Inst 2005;17:193–202.Atsu N, Ekici S, Oge OO, Ergen A, Hascelik G, OzenH. False-positive results of the NMP22 test due tohematuria. J Urol 2002;167:555–8.Bakhos R, Shankey TV, Flanigan RC, Fisher S, WojcikEM. Comparative analysis of DNA flow cytometry andcytology of bladder washings: review of discordant cases.Diagn Cytopathol 2000;22:65–9.Baltaci S, Suzer O, Ozer G, Beduk Y, Gou O. The efficacyof urinary cytology in the detection of recurrent bladdertumours. Int Urol Nephrol 1996;28:649–53.Baron A, Mastroeni F, Moore PS, Bonetti F, OrlandiniS, Manfrin E, et al. Detection of bladder cancer by semiautomatedmicrosatellite analysis of urine sediment. AdvClin Pathol 2000;4:19–24.Bartlett JMS, White A, Stewart CJR, McNicol AM,Underwood MA. Molecular genetic screening for urinedetection of bladder cancer. UroOncology 2002;2:81–5.Bartoletti R, Dal Canto M, Cai T, Piazzini M, TravagliniF, Gavazzi A, et al. Early diagnosis and monitoring ofsuperficial transitional cell carcinoma by microsatelliteanalysis on urine sediment. Oncol Rep 2005;13:531–7.Bass RA, Hemstreet GP, III, Honker NA, Hurst RE,Doggett RS. DNA cytometry and cytology by quantitativefluorescence image analysis in symptomatic bladdercancer patients. Int J Cancer 1987;40:698–705.Bialkowska-Hobrzanska H, Bowles L, Bukala B, JosephMG, Fletcher R, Razvi H. Comparison of humantelomerase reverse transcriptase messenger RNA andtelomerase activity as urine markers for diagnosis ofbladder carcinoma. Mol Diagn 2000;5:267–77.Bianco FJ, Jr, Gervasi DC, Tiguert R, Grignon DJ,Pontes JE, Crissman JD, et al. Matrix metalloproteinase-9expression in bladder washes from bladder cancerpatients predicts pathological stage and grade. ClinCancer Res 1998;4:3011–6.Bollmann M, Heller H, Bankfalvi A, Griefingholt H,Bollmann R. Quantitative molecular urinary cytologyby fluorescence in situ hybridization: a tool for tailoringsurveillance of patients with superficial bladder cancer?BJU Int 2005;95:1219–25.Bowles L, Bialkowska-Hobrzanska H, Bukala B, Nott L,Razvi H. A prospective evaluation of the diagnostic andpotential prognostic utility of urinary human telomerasereverse transcriptase mRNA in patients with bladdercancer. Can J Urol 2004;11:2438–44.Cajulis RS, Haines GK, III, Frias-Hidvegi D, McVaryK. Interphase cytogenetics as an adjunct in thecytodiagnosis of urinary bladder carcinoma. Acomparative study of cytology, flow cytometry andinterphase cytogenetics in bladder washes. Anal QuantCytol Histol 1994;16:1–10.Cajulis RS, Haines GK, III, Frias-Hidvegi D, McVaryK, Bacus JW. Cytology, flow cytometry, image analysis,and interphase cytogenetics by fluorescence in situhybridization in the diagnosis of transitional cellcarcinoma in bladder washes: a comparative study. DiagnCytopathol 1995;13:214–23.Chan MW, Chan LW, Tang NL, Tong JH, Lo KW, LeeTL, et al. Hypermethylation of multiple genes in tumortissues and voided urine in urinary bladder cancerpatients. Clin Cancer Res 2002;8:464–70.Chang SH, Kim YH, Kim ME. Urinary survivin testcompared to the nuclear matrix protein (NMP)-22 testand urine cytology for the diagnosis of bladder cancer.Korean J Urol 2006;47:1041–5.Chen GL, El Gabry EA, Bagley DH. Surveillance ofupper urinary tract transitional cell carcinoma: the roleof ureteroscopy, retrograde pyelography, cytology andurinalysis. J Urol 2000;164:1901–4.Cheng L, Reiter RE, Jin Y, Sharon H, Wieder J, LaneTF, et al. Immunocytochemical analysis of prostate stemcell antigen as adjunct marker for detection of urothelialtransitional cell carcinoma in voided urine specimens. JUrol 2003;169:2094–100.Chow NH, Tzai TS, Cheng HL, Chan SH, Lin JS.Urinary cytodiagnosis: can it have a different prognosticimplication than a diagnostic test? Urol Int 1994;53:18–23.207© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Appendix 11208Constantinou M, Binka-Kowalska A, Borkowska E, ZajacE, Jalmuzna P, Matych J, et al. Application of multiplexFISH, CGH and MSSCP techniques for cytogeneticand molecular analysis of transitional cell carcinoma(TCC) cells in voided urine specimens. J Applied Genet2006;47:273–5.Dal Canto M, Bartoletti R, Travaglini F, Piazzini M,Lodovichi G, Rizzo M, et al. Molecular urinary sedimentanalysis in patients with transitional cell bladdercarcinoma. Anticancer Res 2003;23:5095–100.Dalbagni G, Han W, Zhang ZF, Cordon-Cardo C, SaigoP, Fair WR, et al. Evaluation of the telomeric repeatamplification protocol (TRAP) assay for telomerase asa diagnostic modality in recurrent bladder cancer. ClinCancer Res 1997;3:1593–8.Daniely M, Rona R, Kaplan T, Olsfanger S, Elboim L,Zilberstien Y, et al. Combined analysis of morphologyand fluorescence in situ hybridization significantlyincreases accuracy of bladder cancer detection in voidedurine samples. Urology 2005;66:1354–9.Desgrippes A, Izadifar V, Assailly J, Fontaine E,Beurton D. Diagnosis and prediction of recurrenceand progression in superficial bladder cancers withDNA image cytometry and urinary cytology. BJU Int2000;85:434–6.D’Hallewin M-A, Baert L. Initial evaluation of thebladder tumor antigen test in superficial bladder cancer.J Urol 1996;155:475–6.Di Carlo A, Terracciano D, Mariano A, Macchia V.Urinary gelatinase activities (matrix metalloproteinases2 and 9) in human bladder tumors. Oncol Rep2006;15:1321–6.Feil G, Zumbragel A, Paulgen-Nelde HJ, HennenlotterJ, Maurer S, Krause S, et al. Accuracy of the ImmunoCytassay in the diagnosis of transitional cell carcinoma ofthe urinary bladder. Anticancer Res 2003;23:963–7.Fornari D, Steven K, Hansen AB, Vibits H, JepsenJV, Poulsen AL, et al. Microsatellite analysis of urinesediment versus urine cytology for diagnosingtransitional cell tumors of the urinary bladder. APMIS2004;112:148–52.Frau DV, Usai P, Dettori T, Caria P, De Lisa A, VanniR. Fluorescence in situ hybridization patterns in newlydiagnosed superficial bladder lesions and correspondingbladder washings. Cancer Genet Cytogenet 2006;169:21–6.Friedrich MG, Hellstern A, Toma MI, Hammerer P,Huland H. Are false-positive urine markers for thedetection of bladder carcinoma really wrong or do theypredict tumor recurrence? Eur Urol 2003;43:146–50.Frigerio S, Padberg BC, Strebel RT, Lenggenhager DM,Messthaler A, Abdou MT, et al. Improved detection ofbladder carcinoma cells in voided urine by standardizedmicrosatellite analysis. Int J Cancer 2007;121:329–38.Gibanel R, Ribal MJ, Filella X, Ballesta AM, MolinaR, Alcaraz A, et al. BTA TRAK urine test increasesthe efficacy of cytology in the diagnosis of low-gradetransitional cell carcinoma of the bladder. Anticancer Res2002;22:1157–60.Gilbert SM, Veltri RW, Sawczuk A, Shabsigh A, KnowlesDR, Bright S, et al. Evaluation of DD23 as a marker fordetection of recurrent transitional cell carcinoma of thebladder in patients with a history of bladder cancer.Urology 2003;61:539–43.Godekmerdan A, Yahsi S, Semercioz A, Ilhan F, AkpolatN, Yekeler H. Determination of nuclear matrix protein22 levels in cystitis, urothelial dysplasia and urothelialcarcinoma. Turk J Med Sci 2006;36:93–6.Halling KC, King W, Sokolova IA, Karnes RJ, Meyer RG,Powell EL, et al. A comparison of BTA stat, hemoglobindipstick, telomerase and Vysis UroVysion assays forthe detection of urothelial carcinoma in urine. J Urol2002;167:2001–6.Hautmann S, Toma M, Lorenzo Gomez MF, FriedrichMG, Jaekel T, Michl U, et al. Immunocyt and the HA-HAase urine tests for the detection of bladder cancer: aside-by-side comparison. Eur Urol 2004;46:466–71.Hoshi S, Takahashi T, Satoh M, Numahata K, SuzukiK-I, Ohyama C, et al. Telomerase activity. Simplificationof assay and detection in bladder tumor and urinaryexfoliated cells. Urol Oncol 2000;5:25–30.Hwang EC, Park SH, Jung SI, Kwon DD, Park K, RyuSB, et al. Usefulness of liquid-based preparation in urinecytology. Intl J Urol 2007;14:626–9.Ianari A, Sternberg CN, Rossetti A, Van Rijn A, DeiddaA, Giannarelli D, et al. Results of Bard BTA test inmonitoring patients with a history of transitional cellcancer of the bladder. Urology 1997;49:786–9.Inoue T, Nasu Y, Tsushima T, Miyaji Y, Murakami T,Kumon H. Chromosomal numerical aberrations ofexfoliated cells in the urine detected by fluorescence insitu hybridization: clinical implication for the detectionof bladder cancer. Urol Res 2000;28:57–61.Ishiwata S, Takahashi S, Homma Y, Tanaka Y, KameyamaS, Hosaka Y, et al. Noninvasive detection and predictionof bladder cancer by fluorescence in situ hybridizationanalysis of exfoliated urothelial cells in voided urine.Urology 2001;57:811–15.Kang JU, Koo SH, Jeong TE, Kwon KC, Park JW, JeonCH. Multitarget fluorescence in situ hybridization andmelanoma antigen genes analysis in primary bladdercarcinoma. Cancer Genet Cytogenet 2006;164:32–8.

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Katz RL, Sinkre PA, Zhang HH, Kidd L, Johnston D.Clinical significance of negative and equivocal urinarybladder cytology alone and in combination with DNAimage analysis and cystoscopy. Cancer 1997;81:354–64.Kibar Y, Goktas S, Kilic S, Yaman H, Onguru O, PekerAF. Prognostic value of cytology, nuclear matrix protein22 (NMP22) test, and urinary bladder cancer II (UBC II)test in early recurrent transitional cell carcinoma of thebladder. Ann Clin Lab Sci 2006;36:31–8.Kinoshita H, Ogawa O, Kakehi Y, Mishina M, MitsumoriK, Itoh N, et al. Detection of telomerase activity inexfoliated cells in urine from patients with bladdercancer. J Natl Cancer Inst 1997;89:724–30.Kipp BR, Karnes RJ, Brankley SM, Harwood AR,Pankratz VS, Sebo TJ, et al. Monitoring intravesicaltherapy for superficial bladder cancer using fluorescencein situ hybridization. J Urol 2005;173:401–4.Kirollos MM, McDermott S, Bradbrook RA. Theperformance characteristics of the bladder tumourantigen test. Br J Urol 1997;80:30–4.Konety BR, Metro MJ, Melham MF, Salup RR.Diagnostic value of voided urine and bladder barbotagecytology in detecting transitional cell carcinoma of theurinary tract. Urol Int 1999;62:26–30.Krause FS, Rauch A, Schrott KM, Engehausen DG.Clinical decisions for treatment of different stagedbladder cancer based on multitarget fluorescence in situhybridization assays? World J Urol 2006;24:418–22.Krupski T, Moskaluk C, Boyd JC, TheodorescuD. A prospective pilot evaluation of urinary andimmunohistochemical markers as predictors of clinicalstage of urothelial carcinoma of the bladder. BJU Int2000;85:1027–32.Kumar NU, Dey P, Mondal AK, Singh SK, Vohra H.DNA flow cytometry and bladder irrigation cytologyin detection of bladder carcinoma. Diagn Cytopathol2001;24:153–6.Landman J, Chang Y, Kavaler E, Droller MJ, Liu BC.Sensitivity and specificity of NMP-22, telomerase, andBTA in the detection of human bladder cancer. Urology1998;52:398–402.Larsson PC, Beheshti B, Sampson HA, Jewett MA,Shipman R. Allelic deletion fingerprinting of urine cellsediments in bladder cancer. Mol Diagn 2001;6:181–8.Laudadio J, Keane TE, Reeves HM, Savage SJ, HodaRS, Lage JM, et al. Fluorescence in situ hybridization fordetecting transitional cell carcinoma: implications forclinical practice. BJU Int 2005;96:1280–5.Lee DH, Yang SC, Hong SJ, Chung BH, Kim IY.Telomerase: a potential marker of bladder transitional© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.cell carcinoma in bladder washes. Clin Cancer Res1998;4:535–8.Lekili M, Sener E, Demir MA, Temelta G, Muezzinolu T,Buyuksu C. Comparison of the nuclear matrix protein 22with voided urine cytology in the diagnosis of transitionalcell carcinoma of the bladder. Urol Res 2004;32:124–8.Lubbe L, Nowack R, May M, Ullmann K, Gunia S,Kaufmann O, et al. FISH – a new noninvasive methodfor the diagnosis of urinary bladder carcinomas. Clin Lab2004;50:395–402.Mack D, Frick J. Diagnostic problems of urine cytologyon initial follow-up after intravesical immunotherapywith Calmette-Guérin bacillus for superficial bladdercancer. Urol Int 1994;52:204–7.Mao L, Schoenberg MP, Scicchitano M, Erozan YS, MerloA, Schwab D, Sidransky D, et al. Molecular detection ofprimary bladder cancer by microsatellite analysis. Science1996;271:659–62.Marin-Aguilera M, Mengual L, Ribal MJ, Burset M,Arce Y, Ars E, et al. Utility of a multiprobe fluorescencein situ hybridization assay in the detection of superficialurothelial bladder cancer. Cancer Genet Cytogenet2007;173:131–5.Mattioli S, Seregni E, Caperna L, Botti C, Savelli G,Bombardieri E. BTA-TRAK combined with urinarycytology is a reliable urinary indicator of recurrenttransitional cell carcinoma (TCC) of the bladder. Int JBiol Markers 2000;15:219–25.Meloni AM, Peier AM, Haddad FS, Powell IJ, BlockAW, Huben RP, et al. A new approach in the diagnosisand follow-up of bladder cancer. FISH analysis of urine,bladder washings, and tumors. Cancer Genet Cytogenet1993;71:105–18.Menendez V, Filella X, Alcover JA, Molina R,Mallafre JM, Ballesta AM, et al. Usefulness of urinarynuclear matrix protein 22 (NMP22) as a marker fortransitional cell carcinoma of the bladder. Anticancer Res2000;20:1169–72.Menendez V, Fernandez-Suarez A, Galan JA, Perez M,Garcia-Lopez F. Diagnosis of bladder cancer by analysisof urinary fibronectin. Urology 2005;65:284–9.Mezzelani A, Dagrada G, Alasio L, Sozzi G, Pilotti S.Detection of bladder cancer by multitarget multicolourFISH: comparative analysis on archival cytology andparaffin-embedded tissue. Cytopathology 2002;13:317–25.Mian C, Lodde M, Comploj E, Palermo S, Mian M,Maier K, et al. The value of the ImmunoCyt/uCyt+ testin the detection and follow-up of carcinoma in situ of theurinary bladder. Anticancer Res 2005;25:3641–4.209

Appendix 11210Minimo C, Tawfiek ER, Bagley DH, McCue PA,Bibbo M. Grading of upper urinary tract transitionalcell carcinoma by computed DNA content and p53expression. Urology 1997;50:869–74.Misra V, Gupta SC, Tandon SP, Gupta AK, Sircar S.Cytohistological study of urinary bladder neoplasms. IndJ Pathol Microbiol 2000;43:303–9.Mora LB, Nicosia SV, Pow-Sang JM, Ku NK, Diaz JI,Lockhart J, et al. Ancillary techniques in the followup oftransitional cell carcinoma: a comparison of cytology,histology and deoxyribonucleic acid image analysiscytometry in 91 patients. J Urol 1996;156:49–55.Mungan NA, Kulacoglu S, Basar M, Sahin M, Witjes JA.Can sensitivity of voided urinary cytology or bladderwash cytology be improved by the use of different urinaryportions? Urol Int 1999;62:209–12.Oge O, Atsu N, Sahin A, Ozen H. Comparison of BTAstat and NMP22 tests in the detection of bladder cancer.Scand J Urol Nephrol 2000;34:349–51.Paez A, Coba JM, Murillo N, Fernandez P, de la CalMA, Lujan M, et al. Reliability of the routine cytologicaldiagnosis in bladder cancer. Eur Urol 1999;35:228–32.Pajor G, Sule N, Alpar D, Kajtar B, Kneif M, BollmannD, et al. Increased efficiency of detecting geneticallyaberrant cells by UroVysion test on voided urinespecimens using automated immunophenotypicalpreselection of uroepithelial cells. Cytometry A2008;73:259–65.Panani AD, Kozirakis D, Anastasiou J, Babanaraki A,Malovrouvas D, Roussos C. Is aneusomy of chromosome9 alone a valid biomarker for urinary bladder cancerscreening? Anticancer Res 2006;26:1161–5.Paoluzzi M, Cuttano MG, Mugnaini P, Salsano F,Giannotti P. Urinary dosage of nuclear matrix protein22 (NMP22) like biologic marker of transitional cellcarcinoma (TCC): a study on patients with hematuria.Arch Ital Urol Androl 1999;71:13–18.Placer J, Espinet B, Salido M, Sole F, Gelabert-Mas A.Clinical utility of a multiprobe FISH assay in voidedurine specimens for the detection of bladder cancer andits recurrences, compared with urinary cytology. Eur Urol2002;42:547–52.Planz B, Striepecke E, Jakse G, Bocking A. Use of LewisX antigen and deoxyribonucleic acid image cytometry toincrease sensitivity of urinary cytology in transitional cellcarcinoma of the bladder. J Urol 1998;159:384–7.Planz B, Synek C, Robben J, Bocking A, MarbergerM. Diagnostic accuracy of DNA image cytometry andurinary cytology with cells from voided urine in thedetection of bladder cancer. Urology 2000;56:782–6.Planz B, Synek C, Deix T, Bocking A, Marberger M.Diagnosis of bladder cancer with urinary cytology,immunocytology and DNA-image-cytometry. Anal CellPathol 2001;22:103–9.Pu YS, Tsai TC, Hsieh TS, Huang HH, Kuo SH,Hsueh WC. Role of urinary cytology and urinarydeoxyribonucleic acid flow cytometry in the diagnosis ofbladder cancer. J Formos Med Assoc 1994;93:216–21.Raab SS, Lenel JC, Cohen MB. Low grade transitionalcell carcinoma of the bladder. Cytologic diagnosis bykey features as identified by logistic regression analysis.Cancer 1994;74:1621–6.Raab SS, Slagel DD, Jensen CS, Teague MW, Savell VH,Ozkutlu D, et al. Transitional cell carcinoma: cytologiccriteria to improve diagnostic accuracy. Modern Pathol1996;9:225–32.Raica M, Zylis D, Cimpean AM. Cytokeratin 20,34betaE12 and overexpression of HER-2/neu in urinecytology as predictors of recurrences in superficialurothelial carcinoma. Romanian J Morphol Embryol2005;46:11–15.Rieger-Christ KM, Mourtzinos A, Lee PJ, Zagha RM,Cain J, Silverman M, et al. Identification of fibroblastgrowth factor receptor 3 mutations in urine sedimentDNA samples complements cytology in bladder tumordetection. Cancer 2003;98:737–44.Sadek S, Soloway MS, Hook S, Civantos F. The valueof upper tract cytology after transurethral resection ofbladder tumor in patients with bladder transitional cellcancer. J Urol 1999;161:77–9.Sagerman PM, Saigo PE, Sheinfeld J, CharitonowicsE, Cordon-Cardo C. Enhanced detection of bladdercancer in urine cytology with Lewis X, M344 and 19A211antigens. Acta Cytol 1994;38:517–23.Sanchez-Carbayo M, Urrutia M, Romani R, HerreroM, Gonzalez de Buitrago JM, Navajo JA. Serial urinaryIL-2, IL-6, IL-8, TNFalpha, UBC, CYFRA 21–1 andNMP22 during follow-up of patients with bladder cancerreceiving intravesical BCG. Anticancer Res 2001;21:3041–7.Sawczuk IS, Bagiella E, Sawczuk AT, Yun EJ. Clinicalapplication of NMP22 in the management of transitionalcell carcinoma of the bladder. Cancer Detect Prevent2000;24:364–8.SchmittConrad M, Fornara P, Bohle A, Knipper A,Jocham D. Clinical experience with immunocytologyusing monoclonal antibody 486 P (Medaquic(R)) indetection of bladder cancer. Aktuelle Urol 1997;28:29–33.Selli C, Travaglini F, Dal Canto M, Piazzini M, FronteraS, Bartoletti R. Microsatellite analysis in the detection of

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4bladder carcinoma: a preliminary report. Acta Urol Ital1999;13:117–20.Seripa D, Parrella P, Gallucci M, Gravina C, Papa S,Fortunato P, et al. Sensitive detection of transitional cellcarcinoma of the bladder by microsatellite analysis ofcells exfoliated in urine. Int J Cancer 2001;95:364–9.Serretta V, Vasile P, Gange E, Lino E, Esposito E,Menozzi I. The NMP22 (nuclear matrix protein) testfor surveillance of superficial bladder transitional cellcarcinoma. Experience in 84 patients. Acta Urol Ital1998;12:201–3.Sharkey FE, Sarosdy MF. The significance of centralpathology review in clinical studies of transitional cellcarcinoma in situ. J Urol 1997;157:68–71.Soloway MS, Briggman V, Carpinito GA, Chodak GW,Church PA, Lamm DL, et al. Use of a new tumor marker,urinary NMP22, in the detection of occult or rapidlyrecurring transitional cell carcinoma of the urinary tractfollowing surgical treatment. J Urol 1996;156:363–7.Song Z, Sun X, Wu D. Significance of flow cytometry inthe diagnosis and treatment of bladder tumors. Chin MedSci J 1995;10:38–41.Su CK, Yang CR, Horng YY, Kao YL, Ho HC, Ou YC, etal. NMP22 in transitional cell carcinoma of the urinarybladder. J Chin Med Assoc 2003;66:294–8.Sumi S, Arai K, Kitahara S, Yoshida KI. Preliminaryreport of the clinical performance of a new urinarybladder cancer antigen test: comparison to voided urinecytology in the detection of transitional cell carcinoma ofthe bladder. Clin Chim Acta 2000;296:111–20.Takeuchi Y, Sawada Y, Yabuki D, Masuda E, Satou D,Kuroda K, et al. Clinical study of urine NMP 22 (nuclearmatrix protein 22) as a tumor marker in urinaryepithelial cancer. Aktuelle Urol 2003;34:265–6.Tang CS, Tang CM, Lau YY, Kung IT. Alcoholic carbowaxprefixation and formal alcohol fixation. A new techniquefor urine cytology. Acta Cytol 1997;41:1183–8.Trott PA, Edwards L. Comparison of bladder washingsand urine cytology in the diagnosis of bladder cancer. JUrol 1973;110:664–6.van Rhijn BW, Lurkin I, Kirkels WJ, van der Kwast TH,Zwarthoff EC. Microsatellite analysis – DNA test in urinecompetes with cystoscopy in follow-up of superficialbladder carcinoma: a phase II trial. Cancer 2001;92:768–75.van Rhijn BW, Smit M, van Geenen D, Wijnmaalen A,Kirkels WJ, van der Kwast TH, et al. Surveillance withmicrosatellite analysis of urine in bladder cancer patientstreated by radiotherapy. Eur Urol 2003;43:369–73.© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.Varella-Garcia M, Akduman B, Sunpaweravong P, DiMaria MV, Crawford ED. The UroVysion fluorescence insitu hybridization assay is an effective tool for monitoringrecurrence of bladder cancer. Urol Oncol 2004;22:16–19.Veeramachaneni R, Nordberg ML, Shi R, Herrera GA,Turbat-Herrera EA. Evaluation of fluorescence in situhybridization as an ancillary tool to urine cytologyin diagnosing urothelial carcinoma. Diagn Cytopathol2003;28:301–7.Vriesema JL, Atsma F, Kiemeney LA, Peelen WP, WitjesJA, Schalken JA. Diagnostic efficacy of the ImmunoCyttest to detect superficial bladder cancer recurrence.Urology 2001;58:367–71.Walsh IK, Keane PF, Ishak LM, Flessland KA. The BTAstat test: a tumor marker for the detection of upper tracttransitional cell carcinoma. Urology 2001;58:532–5.Weikert S, Christoph F, Schrader M, Krause H, MillerK, Muller M. Quantitative analysis of survivin mRNAexpression in urine and tumor tissue of bladder cancerpatients and its potential relevance for disease detectionand prognosis. Int J Cancer 2005;116:100–4.Werner W, Ebert W, Wunderlich H, Schubert J, JunkerK. Fluorescence in situ hybridization in the diagnosis oftransitional cell carcinoma. Onkologie 1999;22:216–20.Whisnant RE, Bastacky SI, Ohori NP. Cytologic diagnosisof low-grade papillary urothelial neoplasms (lowmalignant potential and low-grade carcinoma) in thecontext of the 1998 WHO/ISUP classification. DiagnCytopathol 2003;28:186–90.Witjes JA, van der Poel HG, van Balken MR, DebruyneFM, Schalken JA. Urinary NMP22 and karyometry inthe diagnosis and follow-up of patients with superficialbladder cancer. Eur Urol 1998;33:387–91.Wolman SR, Goldman B, Slovak ML, Tangen C, PersonsDL, Wood D. Aneusomy for detection of bladder cancerrecurrence: a Southwest Oncology Group study. CancerGenet Cytogenet 2007;176:22–7.Xu K, Tam PC, Hou S, Wang X, Bai W. The role ofnuclear matrix protein 22 combined with bladder tumorantigen stat test in surveillance of recurring bladdercancer. Chin Med J (Engl) 2002;115:1736–8.Yang SC, Lee DH, Hong SJ, Chung BH, Kim IY.Telomerase activity: a potential marker of bladdertransitional cell carcinoma in bladder washes. Yonsei MedJ 1997;38:155–9.Yokota K, Kanda K, Inoue Y, Kanayama H, KagawaS. Semi-quantitative analysis of telomerase activityin exfoliated human urothelial cells and bladdertransitional cell carcinoma. Br J Urol 1998;82:727–32.211

Appendix 11212Zancan M, Franceschini R, Mimmo C, Vianello M, DiTonno F, Mazzariol C, et al. Free DNA in urine: a newmarker for bladder cancer? Preliminary data. Int J BiolMarkers 2005;20:134–6.Zargar M, Soleimani M, Moslemi M. Comparativeevaluation of urinary bladder cancer antigen and urinecytology in the diagnosis of bladder cancer. Urol J2005;2:137–40.Zhang FF, Arber DA, Wilson TG, Kawachi MH, SlovakML. Toward the validation of aneusomy detection byfluorescence in situ hybridization in bladder cancer:comparative analysis with cytology, cytogenetics, andclinical features predicts recurrence and defines clinicaltesting limitations. Clin Cancer Res 1997;3:2317–28.Zhang J, Zheng S, Fan Z, Gao Y, Di X, Wang D, et al. Acomparison between microsatellite analysis and cytologyof urine for the detection of bladder cancer. Cancer Lett2001;172:55–8.Ziaee SA, Moula SJ, Hosseini Moghaddam SM,Eskandar-Shiri D. Diagnosis of bladder cancer by urinesurvivin, an inhibitor of apoptosis: a preliminary report.Urol J 2006;3:150–3.Required test(s) not reported(n = 79)The use of the bladder-tumour associated analyte testto determine the type of cystoscopy in the follow-up ofpatients with bladder cancer. The United Kingdom andEire Bladder Tumour Antigen Study Group. Br J Urol1997;79:362–6.Akao T, Kakehi Y, Wu X-X, Kinoshita H, Takahashi T,Ogawa O, et al. Semi-quantitative analysis of telomeraseactivity of exfoliated cells in urine of patients withurothelial cancers: causative factors affecting sensitivityand specificity. Urol Oncol 1998;3:118–24.Attallah AM, Sakr HA, Ismail H, Abdel-Hady E, ElDosoky I. An office-based immunodiagnostic assay fordetecting urinary nuclear matrix protein 52 in patientswith bladder cancer. BJU Int 2005;96:334–9.Babjuk M, Kostirova M, Mudra K, Pecher S, SmolovaH, Pecen L, et al. Qualitative and quantitative detectionof urinary human complement factor H-related protein(BTA stat and BTA TRAK) and fragments of cytokeratins8, 18 (UBC rapid and UBC IRMA) as markers fortransitional cell carcinoma of the bladder. Eur Urol2002;41:34–9.Barlandas-Rendon E, Muller MM, Garcia-Latorre E,Heinschink A. Comparison of urine cell characteristicsby flow cytometry and cytology in patients suspected ofhaving bladder cancer. Clin Chem Lab Med 2002;40:817–23.Bian W, Xu Z. Combined assay of CYFRA21–1,telomerase and vascular endothelial growth factor in thedetection of bladder transitional cell carcinoma. Int JUrol 2007;14:108–11.Billerey C, Lamy B, Bittard H, Rozan S, Carbillet JP. Flowcytometry versus urinary cytology in the diagnosis andfollow-up of bladder tumors: critical review of a 5-yearexperience. World J Urol 1993;11:156–60.Bonner RB, Hemstreet GP, III, Fradet Y, Rao JY,Min KW, Hurst RE. Bladder cancer risk assessmentwith quantitative fluorescence image analysis oftumor markers in exfoliated bladder cells. Cancer1993;72:2461–9.Casella R, Shariat SF, Monoski MA, Lerner SP. Urinarylevels of urokinase-type plasminogen activator and itsreceptor in the detection of bladder carcinoma. Cancer2002;95:2494–9.Chautard D, Daver A, Bocquillon V, Verriele V, Colls P,Bertrand G, et al. Comparison of the Bard Trak test withvoided urine cytology in the diagnosis and follow-up ofbladder tumors. Eur Urol 2000;38:686–90.Chong TW, Cheng C. The role of the bladder tumourantigen test in the management of gross haematuria.Singapore Med J 1999;40:578–80.Eissa S, Kenawy G, Swellam M, El Fadle AA, Abd El-AalAA, El Ahmady O. Comparison of cytokeratin 20 RNAand angiogenin in voided urine samples as diagnostictools for bladder carcinoma. Clin Biochem 2004;37:803–10.Eissa S, Labib RA, Mourad MS, Kamel K, El AhmadyO. Comparison of telomerase activity and matrixmetalloproteinase-9 in voided urine and bladder washsamples as a useful diagnostic tool for bladder cancer.Eur Urol 2003;44:687–94.Eissa S, Kassim SK, Labib RA, El Khouly IM, GhafferTM, Sadek M, et al. Detection of bladder carcinomaby combined testing of urine for hyaluronidase andcytokeratin 20 RNAs. Cancer 2005;103:1356–62.Eissa S, Swellam M, el Mosallamy H, Mourad MS,Hamdy N, Kamel K, et al. Diagnostic value of urinarymolecular markers in bladder cancer. Anticancer Res2003;23:4347–55.Ellis WJ, Blumenstein BA, Ishak LM, Enfield DL. Clinicalevaluation of the BTA TRAK assay and comparison tovoided urine cytology and the Bard BTA test in patientswith recurrent bladder tumors. The Multi Center StudyGroup. Urology 1997;50:882–7.Fernandez-Gomez J, Rodriguez-Martinez JJ, BarmadahSE, Garcia RJ, Allende DM, Jalon A, et al. UrinaryCYFRA 21.1 is not a useful marker for the detection ofrecurrences in the follow-up of superficial bladder cancer.Eur Urol 2007;51:1267–74.

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Fitzgerald JM, Ramchurren N, Rieger K, Levesque P,Silverman M, Libertino JA, et al. Identification of H-rasmutations in urine sediments complements cytologyin the detection of bladder tumors. J Natl Cancer Inst1995;87:129–33.Fracasso ME, Franceschetti P, Doria D, Talamini G,Bonetti F. DNA breaks as measured by the alkaline cometassay in exfoliated cells as compared to voided urinecytology in the diagnosis of bladder cancer: a study of105 subjects. Mutat Res 2004;564:57–64.Frick J, Aulitzky W, Fuchs D. The value of urinaryneopterin as an immunological parameter in patientswith malignant tumors of the genitourinary tract. Urol Int1985;40:155–9.Golijanin D, Shapiro A, Pode D. Immunostaining ofcytokeratin 20 in cells from voided urine for detection ofbladder cancer. J Urol 2000;164:1922–5.Gourlay W, Chan V, Gilks CB. Screening for urothelialmalignancies by cytologic analysis and flow cytometry ina community urologic practice: a prospective study. ModPathol 1995;8:394–7.Gregoire M, Fradet Y, Meyer F, Tetu B, Bois R, BedardG, et al. Diagnostic accuracy of urinary cytology, anddeoxyribonucleic acid flow cytometry and cytology onbladder washings during followup for bladder tumors. JUrol 1997;157:1660–4.Gutierrez Banos JL, Henar Rebollo RM, Antolin JuarezFM, Garcia BM. Usefulness of the BTA STAT test for thediagnosis of bladder cancer. Urology 2001;57:685–9.Guy L, Savareux L, Molinie V, Botto H, Boiteux J-P,Lebret T. Should bladder biopsies be performedroutinely after bacillus Calmette-Guérin treatmentfor high-risk superficial transitional cell cancer of thebladder? Eur Urol 2006;50:516–20.Hakenberg OW, Fuessel S, Richter K, Froehner M,Oehlschlaeger S, Rathert P, et al. Qualitative andquantitative assessment of urinary cytokeratin 8 and18 fragments compared with voided urine cytology indiagnosis of bladder carcinoma. Urology 2004;64:1121–6.Heino A, Aaltomaa S, Ala-Opas M. BTA test is superiorto voided urine cytology in detecting malignant bladdertumours. Ann Chir Gynaecol 1999;88:304–7.Jayachandran S, Unni Mooppan MM, Wax SH, KimH. The value of urinary fibrin/fibrinogen degradationproducts as tumor markers in urothelial carcinoma. JUrol 1984;132:21–3.Johnston B, Morales A, Emerson L, Lundie M. Rapiddetection of bladder cancer: a comparative study of pointof care tests. J Urol 1997;158:2098–101.Kavaler E, Landman J, Chang Y, Droller MJ, Liu BC.Detecting human bladder carcinoma cells in voidedurine samples by assaying for the presence of telomeraseactivity. Cancer 1998;82:708–14.Khalbuss W, Goodison S. Immunohistochemicaldetection of hTERT in urothelial lesions: a potentialadjunct to urine cytology. Cytojournal 2006;3:18.Leyh H, Mazeman E. Bard BTA test compared withvoided urine cytology in the diagnosis of recurrentbladder cancer. Eur Urol 1997;32:425–8.Leyh H, Hall R, Mazeman E, Blumenstein BA.Comparison of the Bard BTA test with voided urine andbladder wash cytology in the diagnosis and managementof cancer of the bladder. Urology 1997;50:49–53.Leyh H, Marberger M, Conort P, Sternberg C, PansadoroV, Pagano F, et al. Comparison of the BTA stat test withvoided urine cytology and bladder wash cytology in thediagnosis and monitoring of bladder cancer. Eur Urol1999;35:52–6.Liedl T. Flow cytometric DNA/cytokeratin analysisof bladder lavage: methodical aspects and clinicalimplications. Urol Int 1995;54:22–47.Liu J, Katz R, Shin HJ, Johnston DA, Zhang HZ,Caraway NP. Use of mailed urine specimens indiagnosing urothelial carcinoma by cytology and DNAimage analysis. Acta Cytol 2005;49:157–62.Maier U, Simak R, Neuhold N. The clinical-value ofurinary cytology – 12 years of experience with 615patients. J Clin Pathol 1995;48:314–7.Melissourgos N, Kastrinakis NG, Davilas I, FoukasP, Farmakis A, Lykourinas M. Detection of humantelomerase reverse transcriptase mRNA in urine ofpatients with bladder cancer: evaluation of an emergingtumor marker. Urology 2003;62:362–7.Melissourgos ND, Kastrinakis NG, Skolarikos A, PappaM, Vassilakis G, Gorgoulis VG, et al. Cytokeratin-20immunocytology in voided urine exhibits greatersensitivity and reliability than standard cytology in thediagnosis of transitional cell carcinoma of the bladder.Urology 2005;66:536–41.Miyake H, Hara I, Gohji K, Yamanaka K, Arakawa S,Kamidono S. Urinary cytology and competitive reversetranscriptase-polymerase chain reaction analysis of aspecific CD44 variant to detect and monitor bladdercancer. J Urol 1998;160:2004–8.Miyanaga N, Akaza H, Kameyama S, Hachiya T, OzonoS, Kuroda M, et al. Significance of the BTA test inbladder cancer: a multicenter trial. BTA Study GroupJapan. Int J Urol 1997;4:557–60.213© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Appendix 11214Morsi MI, Youssef AI, Hassouna ME, El Sedafi AS,Ghazal AA, Zaher ER. Telomerase activity, cytokeratin20 and cytokeratin 19 in urine cells of bladder cancerpatients. J Egypt Natl Canc Inst 2006;18:82–92.Nasuti JF, Gomella LG, Ismial M, Bibbo M. Utility ofthe BTA stat test kit for bladder cancer screening. DiagnCytopathol 1999;21:27–9.Neves M, Ciofu C, Larousserie F, Fleury J, Sibony M,Flahault A, et al. Prospective evaluation of geneticabnormalities and telomerase expression in exfoliatedurinary cells for bladder cancer detection. J Urol2002;167:1276–81.Pode D, Shapiro A, Wald M, Nativ O, Laufer M, Kaver I.Noninvasive detection of bladder cancer with the BTAstat test. J Urol 1999;161:443–6.Pogacnik A, Us-Krasovec M. Cytomorphology and flowcytometry in monitoring patients treated for bladdercancer; preliminary results. Radiolo Oncol 1997;31:155–7.Priolo G, Gontero P, Martinasso G, Mengozzi G,Formiconi A, Pelucelli G, et al. Bladder tumor antigenassay as compared to voided urine cytology in thediagnosis of bladder cancer. Clin Chim Acta 2001;305:47–53.Quek P, Chin CM, Lim PH. The role of BTA stat inclinical practice. Ann Acad Med Singapore 2002;31:212–6.Radkhah K, Nowroozi MR, Jabalameli P. Sensitivityand specificity of urinary bladder cancer antigen fordiagnosis of bladder tumor; a comparative study withurinary cytology. Acta Medica Iranica 2005;43:169–72.Raitanen MP, Marttila T, Kaasinen E, Rintala E, Aine R,Tammela TL. Sensitivity of human complement factor Hrelated protein (BTA stat) test and voided urine cytologyin the diagnosis of bladder cancer. J Urol 2000;163:1689–92.Raitanen MP, Hellstrom P, Marttila T, Korhonen H, TaljaM, Ervasti J, et al. Effect of intravesical instillations onthe human complement factor H related protein (BTAstat) test. Eur Urol 2001;40:422–6.Raitanen MP, Kaasinen E, Lukkarinen O, Kauppinen R,Viitanen J, Liukkonen T, et al. Analysis of false-positiveBTA STAT test results in patients followed up for bladdercancer. Urology 2001;57:680–4.Raitanen MP, Leppilahti M, Tuhkanen K, Forssel T,Nylund P, Tammela T, et al. Routine follow-up cystoscopyin detection of recurrence in patients being monitoredfor bladder cancer. Ann Chir Gynaecol 2001;90:261–5.Raitanen MP, Marttila T, Nurmi M, Ala-Opas M,Nieminen P, Aine R, et al. Human complement factor Hrelated protein test for monitoring bladder cancer. J Urol2001;165:374–7.Saika T, Tsushima T, Nasu Y, Kumon H, Ohmori H.Epidermal growth factor in urine from patients withbladder cancer. Urol Res 2000;28:230–4.Sanchez-Carbayo M, Urrutia M, Gonzalez de BuitragoJM, Navajo JA. Evaluation of two new urinary tumormarkers: bladder tumor fibronectin and cytokeratin18 for the diagnosis of bladder cancer. Clin Cancer Res2000;6:3585–94.Sanchini MA, Gunelli R, Nanni O, Bravaccini S, FabbriC, Sermasi A, et al. Relevance of urine telomerase in thediagnosis of bladder cancer. JAMA 2005;294:2052–6.Sarosdy MF, deVere White RW, Soloway MS, SheinfeldJ, Hudson MA, Schellhammer PF, et al. Results of amulticenter trial using the BTA test to monitor for anddiagnose recurrent bladder cancer. J Urol 1995;154:379–83.Sarosdy MF, Hudson MA, Ellis WJ, Soloway MS, deVereWR, Sheinfeld J, et al. Improved detection of recurrentbladder cancer using the Bard BTA stat test. Urology1997;50:349–53.Sawczuk IS, Pickens CL, Vasa UR, Ralph DA, NorrisKA, Miller MC, et al. DD23 Biomarker: a prospectiveclinical assessment in routine urinary cytology specimensfrom patients being monitored for TCC. Urol Oncol2002;7:185–90.Schamhart DH, de Reijke TM, van der Poel HG, WitjesJA, de Boer EC, Kurth K, et al. The Bard BTA test: itsmode of action, sensitivity and specificity, compared tocytology of voided urine, in the diagnosis of superficialbladder cancer. Eur Urol 1998;34:99–106.Schmetter BS, Habicht KK, Lamm DL, Morales A,Bander NH, Grossman HB, et al. A multicenter trialevaluation of the fibrin/fibrinogen degradation productstest for detection and monitoring of bladder cancer. JUrol 1997;158:801–5.Schroeder GL, Lorenzo-Gomez MF, Hautmann SH,Friedrich MG, Ekici S, Huland H, et al. A side by sidecomparison of cytology and biomarkers for bladdercancer detection. J Urol 2004;172:1123–6.Shariat SF, Herman MP, Casella R, Lotan Y, Karam JA,Stenman UH. Urinary levels of tumor-associated trypsininhibitor (TATI) in the detection of transitional cellcarcinoma of the urinary bladder. Eur Urol 2005;48:424–31.Shariat SF, Matsumoto K, Casella R, Jian W, Lerner SP.Urinary levels of soluble e-cadherin in the detection oftransitional cell carcinoma of the urinary bladder. EurUrol 2005;48:69–76.Slaton JW, Dinney CPN, Veltri RW, Miller MC, LiebertM, O’Dowd et al. Deoxyribonucleic acid ploidy enhances

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4the cytological prediction of recurrent transitional cellcarcinoma of the bladder. J Urol 1997;158:806–11.Sole M, Alos L, Mallofre C, Romero JA, Muntane J,Cardesa A. Bladder wash flow cytometry in transitionalcell carcinoma: useful or misleading? Urol Res1995;22:361–5.Takashi M, Schenck U, Kissel K, Leyh H, Treiber U.Use of diagnostic categories in urinary cytology incomparison with the bladder tumour antigen (BTA) testin bladder cancer patients. Int Urol Nephrol 1999;31:189–96.Tanaka K, Takashi M, Miyake K, Koshikawa T.Immunocytochemically demonstrated expression ofepithelial membrane antigen and carcinoembryonicantigen by exfoliated urinary cells in patients withbladder cancer. Urol Int 1994;52:140–4.Tauber S, Schneede P, Liedl B, Liesmann F, Zaak D,Hofstetter A. Fluorescence cytology of the urinarybladder. Urology 2003;61:1067–71.Thomas L, Leyh H, Marberger M, Bombardieri E, BassiP, Pagano F, et al. Multicenter trial of the quantitativeBTA TRAK assay in the detection of bladder cancer. ClinChem 1999;45:472–7.Topsakal M, Karadeniz T, Anac M, Donmezer S, BesisikA. Assessment of fibrin-fibrinogen degradation products(Accu-Dx) test in bladder cancer patients. Eur Urol2001;39:287–91.van der Poel HG, Boon ME, van Stratum P, Ooms EC,Wiener H, Debruyne FM, et al. Conventional bladderwash cytology performed by four experts versusquantitative image analysis. Mod Pathol 1997;10:976–82.van der Poel HG, van Balken MR, Schamhart DH,Peelen P, de Reijke T, Debruyne FM, et al. Bladder washcytology, quantitative cytology, and the qualitative BTAtest in patients with superficial bladder cancer. Urology1998;51:44–50.van der Poel HG, van Rhijn BW, Peelen P, Debruyne FM,Boon ME, Schalken JA. Consecutive quantitative cytologyin bladder cancer. Urology 2000;56:584–8.Varaldo M, Moretti M, Malcangi B, Cichero A, PittalugaP. Flow cytometry in the follow-up of superficial bladdercancer. Acta Urol Ital 1995;9:207–9.Weikert S, Krause H, Wolff I, Christoph F, Schrader M,Emrich T, et al. Quantitative evaluation of telomerasesubunits in urine as biomarkers for noninvasive detectionof bladder cancer. Int J Cancer 2005;117:274–80.Wu WJ, Liu LT, Huang CH, Chang SF, Chang LL.Telomerase activity in human bladder tumors andbladder washing specimens. Kaohsiung J Med Sci2001;17:602–9.© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.Zhao J, He D, Yang L, He H, Nan X. The study ofmicrosatellites alteration in diagnoses of bladder cancer.Acad J Xi’an Jiaotong Univ 2006;18:73–7.Required study design not met(n = 14)New clinical data further supports accuracy of the VysisUroVysion test. Expert Rev Mol Diagn 2002;2:93–4.Akhtar M, Gallagher L, Rohan S. Survivin: role indiagnosis, prognosis, and treatment of bladder cancer.Adv Anat Pathol 2006;13:122–6.Barak V, Goike H, Panaretakis KW, Einarsson R. Clinicalutility of cytokeratins as tumor markers. Clin Biochem2004;37:529–40.Clarke NW, Sangar VK. Urine telomerase activity for thediagnosis of bladder cancer: commentary. Nature ClinPract Urol 2006;3:192–3.Delanghe J. New screening diagnostic techniques inurinalysis. Acta Clin Belg 2007;62:155–61.Duffy MJ, O’Donovan N, Brennan DJ, Gallagher WM,Ryan BM. Survivin: a promising tumor biomarker. CancerLett 2007;249:49–60.Ellwein LB. Bladder cancer screening: lessons from abiologically based model of bladder cancer progressionand therapeutic intervention. J Occup Med 1990;32:806–11.Getzenberg RH. Urine-based assays for bladder cancer.Lab Med 2003;34:613–7.Grossman HB. Bladder cancer markers: translatingdiscovery to the clinic. J Clin Ligand Assay 2004;26:184–6.Grossman HB. Fluorescence in situ hybridization fordetecting transitional cell carcinoma: implications forclinical practice. In: Laudadio J, Keane TE, Reeves HM,Savage SJ, Hoda RS, Lage JM, Wolff DJ, Department ofPathology and Laboratory Medicine, University of SouthCarolina, Charleston, SC. Urol Oncol 2006;24:270–1.Master VA, Meng MV, Koppie TM, Carroll PR, GrossfeldGD. Origin of urothelial carcinoma after renal transplantdetermined by fluorescence in situ hybridization. J Urol2002;167:2521–2.Mellon JK. Can nomograms using urinary NMP22predict recurrence and progression of superficial bladdercancer? Commentary. Nature Clin Pract Urol 2005;2:370–1.Moonen PM, Kiemeney LA, Witjes JA. Urinary NMP22BladderChek test in the diagnosis of superficial bladdercancer. Eur Urol 2005;48:951–6.215

Appendix 11216Yossepowitch O, Herr HW, Donat SM. Use of urinarybiomarkers for bladder cancer surveillance: patientperspectives. J Urol 2007;177:1277–82.Required outcomes notreported (n = 13)Boman H, Hedelin H, Jacobsson S, Holmang S. Newlydiagnosed bladder cancer: the relationship of initialsymptoms, degree of microhematuria and tumor markerstatus. J Urol 2002;168:1955–9.Britton JP, Dowell AC, Whelan P, Harris CM. Acommunity study of bladder cancer screening bythe detection of occult urinary bleeding. J Urol1992;148:788–90.Chen Y-T, Hayden CL, Marchand KJ, Makuch RW.Comparison of urine collection methods for evaluatingurinary nuclear matrix protein, NMP22, as a tumormarker. J Urol 1997;158:1899–901.Mahnert B, Tauber S, Kriegmair M, Nagel D,Holdenrieder S, Hofmann K, et al. Measurements ofcomplement factor H-related protein (BTA-TRAK assay)and nuclear matrix protein (NMP22 assay) – usefuldiagnostic tools in the diagnosis of urinary bladdercancer? Clin Chem Lab Med 2003;41:104–10.Mansoor I. Analysis of urine cytology at a communityhospital. JAMC 2003;15:20–3.Murphy WM, Rivera-Ramirez I, Medina CA, WrightNJ, Wajsman Z. The bladder tumor antigen (BTA) testcompared to voided urine cytology in the detection ofbladder neoplasms. J Urol 1997;158:2102–6.Pariente JL, Bordenave L, Michel P, Latapie MJ,Ducassou D, Le Guillou M. Initial evaluation of CYFRA21–1 diagnostic performances as a urinary marker inbladder transitional cell carcinoma. J Urol 1997;158:338–41.Shariat SF, Zippe C, Ludecke G, Boman H, Sanchez-Carbayo M, Casella R, et al. Nomograms includingnuclear matrix protein 22 for prediction of diseaserecurrence and progression in patients with Ta, T1 orCIS transitional cell carcinoma of the bladder. J Urol2005;173:1518–25.Svatek RS, Lee D, Lotan Y. Correlation of office-basedcystoscopy and cytology with histologic diagnosis: howgood is the reference standard? Urology 2005;66:65–8.Toma MI, Friedrich MG, Hautmann SH, Jakel KT,Erbersdobler A, Hellstern A, et al. Comparison of theImmunoCyt test and urinary cytology with other urinetests in the detection and surveillance of bladder cancer.World J Urol 2004;22:145–9.Tut VM, Hildreth AJ, Kumar M, Mellon JK. Does voidedurine cytology have biological significance? Br J Urol1998;82:655–9.Wiener HG, Vooijs GP, Hof-Grootenboer B. Accuracyof urinary cytology in the diagnosis of primary andrecurrent bladder cancer. Acta Cytol 1993;37:163–9.Wiener HG, Remkes GW, Schatzl G, Susani M,Breitenecker G. Quick-staining urinary cytology andbladder wash image analysis with an integrated riskclassification: a worthwhile improvement in the follow-upof bladder cancer? Cancer 1999;87:263–9.Criteria for control group notmet (n = 10)Houskova L, Zemanova Z, Babjuk M, MelichercikovaJ, Pesl M, Michalova K. Molecular cytogeneticcharacterization and diagnostics of bladder cancer.Neoplasma 2007;54:511–16.Junker K, Werner W, Mueller C, Ebert W, Schubert J,Claussen U. Interphase cytogenetic diagnosis of bladdercancer on cells from urine and bladder washing. Int JOncol 1999;14:309–13.Lee MY, Tsou MH, Cheng MH, Chang DS, Yang AL, KoJS. Clinical application of NMP22 and urinary cytologyin patients with hematuria or a history of urothelialcarcinoma. World J Urol 2000;18:401–5.Shiff C, Veltri R, Naples J, Quartey J, Otchere J, AnyanW, et al. Ultrasound verification of bladder damageis associated with known biomarkers of bladdercancer in adults chronically infected with Schistosomahaematobium in Ghana. Trans R Soc Trop Med Hygiene2006;100:847–54.Sun Y, He DL, Ma Q, Wan XY, Zhu GD, Li L, et al.Comparison of seven screening methods in the diagnosisof bladder cancer. Chin Med J (Engl) 2006;119:1763–71.Svatek RS, Karam J, Karakiewicz PI, Gallina A, CasellaR, Roehrborn CG, et al. Role of urinary cathepsin B andL in the detection of bladder urothelial cell carcinoma. JUrol 2008;179:478–84.Svatek RS, Herman MP, Lotan Y, Casella R, Hsieh JT,Sagalowsky AI, et al. Soluble Fas – a promising novelurinary marker for the detection of recurrent superficialbladder cancer. Cancer 2006;106:1701–7.Tsui KH, Chen SM, Wang TM, Juang HH, Chen CL, SunGH, et al. Comparisons of voided urine cytology, nuclearmatrix protein-22 and bladder tumor associated antigentests for bladder cancer of geriatric male patients inTaiwan, China. Asian J Androl 2007;9:711–15.Upendra KN, Dey P, Mondal AK, Singh SK, Vohra H.DNA flow cytometry and bladder irrigation cytologyin detection of bladder carcinoma. Diagn Cytopathol2001;24:153–6.Zellweger T, Benz G, Cathomas G, Mihatsch MJ, SulserT, Gasser TC, et al. Multi-target fluorescence in situ

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4hybridization in bladder washings for prediction ofrecurrent bladder cancer. Int J Cancer 2006;119:1660–5.Required reference standard notmet (n = 3)Alishahi S, Byrne D, Goodman CM, Baxby K.Haematuria investigation based on a standard protocol:emphasis on the diagnosis of urological malignancy. J RColl Surg Edinb 2002;47:422–7.Hofland CA, Mariani AJ. Is cytology required for ahematuria evaluation? J Urol 2004;171:324–6.Sanchini MA, Gunelli R, Nanni O, Bravaccini S, FabbriC, Sermasi A, et al. Relevance of urine telomerase in thediagnosis of bladder cancer. JAMA 2005;294:2052–6.Cytology studies predating thepublication year of the earliestincluded biomarker study (n = 3)Loh CS, Spedding AV, Ashworth MT, Kenyon WE,Desmond AD. The value of exfoliative urine cytologyin combination with flexible cystoscopy in the diagnosisof recurrent transitional cell carcinoma of the urinarybladder. Br J Urol 1996;77:655–8.Renshaw AA, Nappi D, Weinberg DS. Cytology of grade1 papillary transitional cell carcinoma. A comparisonof cytologic, architectural and morphometric criteria incystoscopically obtained urine. Acta Cytol 1996;40:676–82.Sack MJ, Artymyshyn RL, Tomaszewski JE, Gupta PK.Diagnostic-value of bladder wash cytology, with specialreference to low-grade urothelial neoplasms. Acta Cytol1995;39:187–94.217© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Appendix 12Characteristics of the biomarkerand cytology studiesStudy Participants Tests Outcomes summaryAbbate 1998 154Study design: case–controlTime period: NSCountry: ItalyEnrolled: 182; analysed: 135No previous history of BC:NS; history of BC: NSAge (years): mean 63, range41 to 89Sex: NSTests and cut-off used:NMP22, 12 U/mlUnit of analysis: patient(n = 135)Sensitivity: 54%Specificity: 87%Bastacky 1999 165Study design: CC-SD (threecentres)Time period: 1990–4Country: USAEnrolled: 1672; analysed: 743No previous history of BC:752; history of BC: 485Age (years): NSSex: NSTests and cut-off used:cytology (VU or BW),subjective assessmentUnit of analysis: patient(n = 743)Sensitivity: 64%Specificity: 93%Bhuiyan 2003 120Study design: C-SDTime period: NSCountry: Saudi Arabia/USAEnrolled: 233; analysed: 231NMP22, 125 cytologyNo previous history of BC:NS; history of BC: NSAge (years): NSSex: NSTests and cut-off used:NMP22, 3.6 U/ml, ≥ 10 U/ml; cytology (VU), subjectiveassessmentUnit of analysis: specimen(n = 231, NMP22; n = 125,cytology)Sensitivity: 25% (NMP2210 U/ml), 40% (cytology)Specificity: 94% (NMP2210 U/ml), 95% (cytology)Boman 2002 155aStudy design: case–controlTime period: Jan 1998 toNov 1999Country: SwedenEnrolled: 250; specimensanalysed: 297 NMP22, 293cytologyNo previous history of BC:NS; history of BC: 174Age (years): NSSex: NSTests and cut-off used:NMP22, ≥ 4 U/ml; cytology(BW), subjective assessmentUnit of analysis: specimen(n = 297, NMP22; n = 293,cytology)Sensitivity: 54% (NMP22),40% (cytology)Specificity: 68 (NMP22), 93%(cytology)Casella 2000 121,145,146Study design: C-SDTime period: Jan 1997 toJun 1999Country: SwitzerlandEnrolled: 235; analysed: 235NMP22, 200 cytologyNo previous history of BC:NS; history of BC: NSAge (years): mean 72, range37 to 97 (M); mean 69, range23 to 96 (F)Sex: 164 M, 71 FTests and cut-off used:NMP22 ≥ 10 U/ml; cytology(BW), subjective assessmentUnit of analysis: patient(n = 235, NMP22; n = 200,cytology)Sensitivity: 52% (NMP22),53% (cytology)Specificity: 84% (NMP22),90% (cytology)Casetta 2000 122Study design: C-SDTime period: Jan 1997 toDec 1998Country: ItalyEnrolled: 196; analysed: 196No previous history of BC:94; history of BC: 102Age (years): mean 68, nohistory BC; mean 69, historyBC; range NSSex: 170 M, 26 FTests and cut-off used:NMP22 ≥10 U/ml, 11 U/ml, 12 U/ml; cytology (VU),subjective assessmentUnit of analysis: patient(n = 196)Sensitivity: 64% (NMP2210 U/ml), 73% (cytology)Specificity: 63% (NMP2210 U/ml), 80% (cytology)219© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Appendix 12Study Participants Tests Outcomes summaryChahal 2001 166Study design: C-SDTime period: Jan 1998 to Jan2000Country: UKEnrolled: 285; analysed: 285No previous history of BC:NS; history of BC: NSAge (years): mean 62, rangeNSSex: 171 M, 114 FTests and cut-off used:cytology (VU), subjectiveassessmentUnit of analysis: patient(n = 285)Sensitivity: 49%Specificity: 94%Chahal 2001 45Study design: CC-SDTime period: NSCountry: UKEnrolled: 211; analysed: 211No previous history of BC:96; history of BC: 115Age (years): NSSex: NSTests and cut-off used:NMP22 ≥ 10 U/ml; cytology(VU), subjective assessmentUnit of analysis: patient(n = 211)Sensitivity: 33% (NMP22),24% (cytology)Specificity: 92 (NMP22), 97%(cytology)Chang 2004 156aStudy design: case–control(no history of disease)Time period: NSCountry: ChinaEnrolled: 399; analysed: 314No previous history of BC:NS; history of BC: NSAge (years): mean 53, range3 to 91Sex: 220 M, 111 FTests and cut-off used:NMP22 ≥ 7.5 U/mlUnit of analysis: patient(n = 314)Sensitivity: 36%Specificity: 83%Daniely 2007 94Study design: C-SDTime period: 2003–4Country: IsraelEnrolled: 115; analysed: 115No previous history of BC:49; history of BC: 66Age (years): NSSex: 73 M, 42 FTests and cut-off used:FISH, minimum of four cellswith gains of two or morechromosomes or 12 or morecells with homozygous lossof the 9p21 locus + cytologyUnit of analysis: patient(n = 115)Sensitivity: 100%Specificity: 50%Del Nero 1999 123Study design: C-SDTime period: NSCountry: ItalyEnrolled: 105; analysed: 105No previous history of BC: 0;history of BC: 105Age (years): mean 54, range42 to 73Sex: 92 M, 13 FTests and cut-off used:NMP22 ≥ 5 U/ml, 6 U/ml,10 U/ml; cytology (VU),subjective assessmentUnit of analysis: patient(n = 105)Sensitivity: 83% (NMP2210 U/ml), 47% (cytology)Specificity: 87% (NMP22 10U/ml), 83% (cytology)Friedrich 2003 95,96aStudy design: C-SDTime period: NSCountry: GermanyEnrolled: 103; analysed: 103No previous history of BC:55; history of BC: 48Age (years): NSSex: NSTests and cut-off used:NMP22 ≥ 10 U/ml; FISH, 20%of cells had a gain of two ormore chromosomes (3, 7 or17) or 40% of cells had a gainof one chromosome or 40%loss of 9p21 locusUnit of analysis: patient(n = 103)Sensitivity: 70% (NMP22),67% (FISH)Specificity: 65% (NMP22),89% (FISH)Garbar 2007 167Study design: C-SDTime period: 2002–4Country: BelgiumEnrolled: 139; analysed: 139No previous history of BC:NS; history of BC: NSAge (years): mean 69 (men),range NS; mean 68 (female),range NSSex: 90 M, 49 FTests and cut-off used:cytology (BW), subjectiveassessmentUnit of analysis: specimen(n = 592)Sensitivity: 60%Specificity: 95%Giannopoulos 2001 124,125aStudy design: C-SDTime period: NSCountry: GreeceEnrolled: 234; analysed: 213No previous history of BC:118; history of BC: 95Age (years): mean 66, range25 to 93Sex: 200 M, 34 FTests and cut-off used:NMP22 ≥ 8 U/mlUnit of analysis: patient(n = 213)Sensitivity: 64%Specificity: 72%220

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Study Participants Tests Outcomes summaryGrossman 2005 126Study design: CC-SD (23centres)Time period: Sep 2001 toMay 2002Country: USAEnrolled: 1331; analysed: 1331No previous history ofBC:1331; history of BC: 1331Age (years): mean 59, range18 to 96Sex: 759 M, 572 FTests and cut-off used:NMP22 ≥ 10 U/ml; cytology(VU), subjective assessmentUnit of analysis: patient(n = 1331, NMP22; n = 1287,cytology)Sensitivity: 56% (NMP22),16% (cytology)Specificity: 86% (NMP22),99% (cytology)Grossman 2006 127Study design: CC-SD (23centres)Time period: Sep 2001 toFeb 2002Country: USAEnrolled: 668; analysed: 668No previous history of BC: 0;history of BC: 668Age (years): mean 71, range30 to 95Sex: 503 M, 165 FTests and cut-off used:NMP22 ≥ 10 U/ml; cytology(VU), subjective assessmentUnit of analysis: patient(n = 668, NMP22; n = 650,cytology)Sensitivity: 50% (NMP22),12% (cytology)Specificity: 87% (NMP22),97% (cytology)Guttierez Banos 2001 128Study design: C-SDTime period: NSCountry: SpainEnrolled: 150; analysed: 150No previous history of BC:64; history of BC: 86Age (years): mean 68, range20 to 91Sex: NSTests and cut-off used:NMP22 ≥ 6 U/ml, 10 U/ml;cytology (VU), subjectiveassessment; cystoscopy(rigid)Unit of analysis: patient(n = 150)Sensitivity: 76% (NMP2210 U/ml), 70% (cytology),100% (cystoscopy)Specificity: 91% (NMP2210 U/ml), 93% (cytology), 89%(cystoscopy)Hakenberg 2000 168Study design: C-SDTime period: Jun 1996 toDec 1997Country: GermanyEnrolled: 374; analysed: 374No previous history of BC:374; history of BC: 0Age (years): mean 68 (men),74 (female), range NSSex: 276 M, 98 FTests and cut-off used:cytology (VU), subjectiveassessmentUnit of analysis: specimen(n = 417)Sensitivity: 76%Specificity: 80%Halling 2000 97Study design: C-SDTime period: NSCountry: USAEnrolled: 265; analysed: 118No previous history of BC:115; history of BC: 150Age (years): mean 70, range36 to 94Sex: 200 M, 65 FTests and cut-off used: FISHfive or more cells polysomy;cytology (VU), subjectiveassessmentUnit of analysis: patient(n = 151, FISH; n = 118,cytology)Sensitivity: 81% (FISH), 58%(cytology)Specificity: 96% (FISH), 98%(cytology)Hughes 1999 129aStudy design: C-SDTime period: NSCountry: USAEnrolled: 107; analysed: 107No previous history of BC: 0;history of BC: 107Age (years): mean 66, range33 to 86Sex: 84 M, 23 FTests and cut-off used:NMP22 ≥ 6.4 U/ml; cytology(VU), subjective assessmentUnit of analysis: specimen(n = 128)Sensitivity: 47% (NMP22),60% (cytology)Specificity: 79% (NMP22),58% (cytology)Junker 2006 98Study design: C-SDTime period: NSCountry: GermanyEnrolled: 141; analysed: 121FISH, 109 cytologyNo previous history of BC:NS; history of BC: NSAge (years): NSSex: NSTests and cut-off used: FISH,five or more cells showedgains of more than onechromosome (3, 7 or 17),or 10 or more cells showedgains of a single chromosome(3, 7 or 17) or 10 or morecells showed homozygousloss of 9p21 locus; cytology(NS)Unit of analysis: patient(n = 121, FISH; n = 109,cytology)Sensitivity: 60% (FISH), 24%(cytology)Specificity: 81% (FISH), 91%(cytology)221© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Appendix 12Study Participants Tests Outcomes summaryKarakiewicz 2006 169,170Study design: C-SD (10centres)Time period: NSCountry: AustriaEnrolled: 2686; analysed: 2542No previous history of BC: 0;history of BC: 2542Age (years): mean 65, range18 to 97Sex: 1910 M, 632 FTests and cut-off used:cytology (VU), subjectiveassessmentUnit of analysis: patient(n = 2542)Sensitivity: 45%Specificity: 95%Kipp 2008 99Study design: C-SDTime period: Mar 2006 toMar 2007Country: USAEnrolled: 124; analysed: 124No previous history of BC:41; history of BC: 81Age (years): mean 72, range45 to 89Sex: 103 M, 21 FTests and cut-off used: FISH,four or more cells hadpolysomic signal patterns(gain of two or more ofthe four chromosomes inan individual cell), 10 ormore cells demonstratedtetrasomy (four signalpatterns for all fourprobes) or > 20% of thecells demonstrated 9p21homozygous deletion (loss oftwo 9p21 signals)Unit of analysis: patient(n = 124)Sensitivity: 62%Specificity: 87%Kowalska 2005 130Study design: C-SDTime period: NSCountry: PolandEnrolled: 98; analysed: 98No previous history of BC: 0;history of BC: 98Age (years): mean 67 (male),64 (female), range 36 to 96Sex: 84 M, 14 FTests and cut-off used:NMP22 ≥ 10 U/mlUnit of analysis: patient(n = 98)Sensitivity: 53%Specificity: 46%Kumar 2006 131Study design: C-SDTime period: NSCountry: IndiaEnrolled: 131; analysed: 131No previous history of BC: 0;history of BC: 131Age (years): mean 67, range32 to 91Sex: 117 M, 14 FTests and cut-off used:NMP22 ≥ 10 U/ml; cytology(VU), subjective assessmentUnit of analysis: patient(n = 131)Sensitivity: 85% (NMP22),41% (cytology)Specificity: 78% (NMP22),96% (cytology)Lahme 2001 132,133Study design: C-SDTime period: NSCountry: GermanyEnrolled: 169; analysed: 109No previous history of BC:40; history of BC: 44Age (years): NSSex: NSTests and cut-off used:NMP22 ≥ 10 U/ml; cytology(VU), subjective assessmentUnit of analysis: patient(n = 109)Sensitivity: 63% (NMP22),45% (cytology)Specificity: 61% (NMP22),93% (cytology)Lee 2001 157aStudy design: case–control(nhd)Time period: NSCountry: South KoreaEnrolled: 106; analysed: 106No previous history of BC:NS; history of BC: NSAge (years): mean 60 (cases),62 (control), range 30 to 78Sex: NSTests and cut-off used:NMP22 7.7 U/ml; cytology(VU), subjective assessmentUnit of analysis: patient(n = 106)Sensitivity: 76% (NMP22),56% (cytology)Specificity: 72% (NMP22),89% (cytology)Lodde 2003 109Study design: CC-SDTime period: NSCountry: Austria, ItalyEnrolled: 235; analysed: 225No previous history of BC:98; history of BC: 137Age (years): mean 72, range32 to 86Sex: NSTests and cut-off used:ImmunoCyt, at least onegreen or one red fluorescentcell; cytology (VU), subjectiveassessment ; ImmunoCyt +cytology (VU)Unit of analysis: patient(n = 225)Sensitivity: 87% (ImmunoCyt),41% (cytology), 90%(ImmunoCyt + cytology)Specificity: 67% (ImmunoCyt),94% (cytology), 68%(ImmunoCyt + cytology)222

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Study Participants Tests Outcomes summaryLodde 2006 110Study design: CC-SDTime period: NSCountry: Austria, ItalyEnrolled: 216; analysed: 195No previous history of BC: 0;history of BC: 216Age (years): NSSex: NSTests and cut-off used:ImmunoCyt, at least onegreen or one red fluorescentcell; cytology (VU), subjectiveassessment ; ImmunoCyt +cytology (VU)Unit of analysis: tumourrecurrence (n = 334,ImmunoCyt; n = 277,cytology; n = 334, ImmunoCyt+ cytology)Sensitivity: 71% (ImmunoCyt),49% (cytology), 86%(ImmunoCyt + cytology)Specificity: 78% (ImmunoCyt),95% (cytology), 78%(ImmunoCyt + cytology)May 2007 107Study design: case–control(nhd)Time period: NSCountry: GermanyEnrolled: 166; analysed: 166No previous history of BC:62; history of BC: 71Age (years): mean 68, range37 to 90Sex: 139 M, 27 FTests and cut-off used:FISH, gain of two or morechromosomes in five ormore cases per slide, orin cases of isolated gainsof chromosome 3, 7 or 17when the proportion of cellswith such a gain was 10% ormore of at least 100 cellsevaluated, or when therewere 10 or more cells with9p21 loss; cytology (VU),subjective assessmentUnit of analysis: patient(n = 166)Sensitivity: 53% (FISH), 71%(cytology)Specificity: 74% (FISH), 84%(cytology)Meiers 2007 100Study design: C-SDTime period: NSCountry: USA, BelgiumEnrolled: 624; analysed: 624No previous history of BC:NS; history of BC: NSAge (years): NSSex: NSTests and cut-off used: FISH,chromosomal gain of two ormore chromosomes (+3, +7,+17) in four or more cellsor deletion of 9p21 in 12 ormore cells; cytology (VU),subjective assessmentUnit of analysis: patient(n = 624)Sensitivity: 93% (FISH), 73%(cytology)Specificity: 90% (FISH), 87%(cytology)Messing 2005 111Study design: C-SD (fourcentres)Time period: Nov 2000 toNov 2003Country: USAEnrolled: 341; analysed: 326No previous history of BC: 0;history of BC: 341Age (years): NSSex: NSTests and cut-off used:ImmunoCyt, one green orone red fluorescent cell;cytology (VU), subjectiveassessment; ImmunoCyt +cytology (VU)Unit of analysis: patient(n = 326)Sensitivity: 81% (ImmunoCyt),23% (cytology), 81%(ImmunoCyt + cytology)Specificity: 75% (ImmunoCyt),93% (cytology), 73%(ImmunoCyt + cytology)Mian 1999 112Study design: CC-SDTime period: Nov 1997 toMar 1998Country: AustriaEnrolled: 264; analysed: 249No previous history of BC:114; history of BC: 150Age (years): mean 66, range21 to 93Sex: 204 M, 60 FTests and cut-off used:ImmunoCyt, one green orone red fluorescent cell;cytology (VU), subjectiveassessment ; ImmunoCyt +cytology (VU)Unit of analysis: patient(n = 249)Sensitivity: 86% (ImmunoCyt),47% (cytology), 90%(ImmunoCyt + cytology)Specificity: 79% (ImmunoCyt),98% (cytology), 79%(ImmunoCyt + cytology)Mian 2000 134Study design: C-SDTime period: NSCountry: AustriaEnrolled: 240; analysed: 240No previous history of BC:81; history of BC: 159Age (years): mean 66, range22 to 92Sex: NSTests and cut-off used:NMP22 ≥ 10 U/mlUnit of analysis: patient(n = 240)Sensitivity: 56%Specificity: 79%223© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Appendix 12Study Participants Tests Outcomes summaryMian 2003 101Study design: CC-SDTime period: NSCountry: Austria, ItalyEnrolled: 181; analysed: 181ImmunoCyt, cytology; 57FISHNo previous history of BC:81; history of BC: 100Age (years): 67, range 32 to83Sex: NSTests and cut-off used:ImmunoCyt, one greenor one red fluorescentcell; FISH, four or moreaneusomic of 25 countedcells; cytology (VU),subjective assessment;ImmunoCyt + cytology (VU)Unit of analysis: patient(n = 181, ImmunoCyt; n = 57,FISH; n = 181, cytology;n = 181, ImmunoCyt +cytology)Sensitivity: 86% (ImmunoCyt),96% (FISH), 45% (cytology),90% (ImmunoCyt + cytology)Specificity: 71% (ImmunoCyt),45% (FISH), 94% (cytology),66% (ImmunoCyt + cytology)Mian 2006 113Study design: C-SDTime period: Jan 2002 to Oct2004Country: ItalyEnrolled: 942; analysed: NSNo previous history of BC: 0;history of BC: 942Age (years): mean 73, range32 to 87Sex: NSTests and cut-off used:ImmunoCyt, one green orone red fluorescent cell;cytology (VU), subjectiveassessment; ImmunoCyt +cytology (VU)Unit of analysis: specimen(n = 1886)Sensitivity: 85% (ImmunoCyt),39% (cytology), 89%(ImmunoCyt + cytology)Specificity: 73% (ImmunoCyt),99% (cytology), 73%(ImmunoCyt + cytology)Miyanaga 1999 135aStudy design: C-SD (13centres)Time period: Aug 1995 toMar 1997Country: JapanEnrolled: 309; analysed: 309No previous history of BC:309; history of BC: 0Age (years): NSSex: 145 M, 164 FTests and cut-off used:NMP22 ≥ 12 U/ml; cytology(VU), subjective assessmentUnit of analysis: patient(n = 309)Sensitivity: 91% (NMP22),55% (cytology)Specificity: 76% (NMP22),100% (cytology)Miyanaga 2003 136aStudy design: C-SDTime period: Jan 2000 to Mar2002Country: JapanEnrolled: 156; analysed: 137No previous history of BC:99; history of BC: 57Age (years): mean 69, range37 to 91Sex: 120 M, 36 FTests and cut-off used:NMP22 ≥ 5 U/ml, 12 U/ml;cytology (VU), subjectiveassessmentUnit of analysis: patient(n = 137)Sensitivity: 19% (NMP2212 U/ml), 7% (cytology)Specificity: 85% (NMP2212 U/ml), 98% (cytology)Moonen 2007 102Study design: C-SDTime period: Mar 2005 toApr 2006Country: the NetherlandsEnrolled: 105; analysed: 95No previous history of BC: 0;history of BC: 105Age (years): mean 70, range44 to 93Sex: 73 M, 22 FTests and cut-off used: FISH,four or more of the 25morphologically abnormalcells showed gains of twoor more chromosomes (3, 7or 17) or 12 or more of the25 cells had no 9p21 signals;cytology (VU), subjectiveassessment; FISH + cytology(VU)Unit of analysis: specimen(n = 103, FISH; n = 108,cytology; n = 103, FISH +cytology)Sensitivity: 39% (FISH), 41%(cytology), 53% (FISH +cytology)Specificity: 90% (FISH), 90%(cytology), 79% (FISH +cytology)Oge 2001 137Study design: C-SDTime period: NSCountry: TurkeyEnrolled: 114; analysed: 76No previous history of BC:37; history of BC: 39Age (years): mean 59 (groups1–3), range 26 to 87Sex: 93 M, 21 FTests and cut-off used:NMP22 ≥ 10 U/mlUnit of analysis: patient(n = 76)Sensitivity: 74%Specificity: 69%Olsson 2001 114Study design: C-SDTime period: Jun 1999 to Jul2000Country: SwedenEnrolled: 121; analysed: 114No previous history of BC:60; history of BC: 61Age (years): mean 68, range15 to 93Sex: 95 M, 26 FTests and cut-off used:ImmunoCyt, one green orone red fluorescent cell;cytology (BW), subjectiveassessmentUnit of analysis: patient(n = 114)Sensitivity: 100%(ImmunoCyt), 58% (cytology)Specificity: 69% (ImmunoCyt),NS (cytology)224

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Study Participants Tests Outcomes summaryOosterhuis 2002 138Study design: C-SDTime period: NSCountry: the NetherlandsEnrolled: 191; analysed: 191No previous history of BC: 0;history of BC: 191Age (years): mean 65, range32 to 89Sex: 146 M, 45 FTests and cut-off used:NMP22 ≥ 10 U/mlUnit of analysis: specimen(n = 431)Sensitivity: 50%Specificity: 68%Parekattil 2003 158aStudy design: case–control(nhd)Time period: Nov 1999 toSep 2000Country: USAEnrolled: 253; analysed: 253No previous history of BC:155; history of BC: 98Age (years): mean 63, range16 to 89Sex: 182 M, 71 FTests and cut-off used:NMP22 ≥ 2.5 U/ml; cytology(VU or BW), subjectiveassessmentUnit of analysis: patient(n = 252, NMP22; n = 253,cytology)Sensitivity: 70% (NMP22),67% (cytology)Specificity: 45% (NMP22),81% (cytology)Piaton 2003 115,116Study design: CC-SD (19centres)Time period: NSCountry: FranceEnrolled: 694; analysed: 651No previous history of BC:236; history of BC: 458Age (years): mean 66, range32 to 92Sex: 550 M, 144 FTests and cut-off used:ImmunoCyt, one green orone red fluorescent cell;cytology (VU), subjectiveassessment ; ImmunoCyt +cytology (VU)Unit of analysis: patient(n = 651, ImmunoCyt; n = 651,cytology; n = 146, ImmunoCyt+ cytology)Sensitivity: 73% (ImmunoCyt),62% (cytology), 82%(ImmunoCyt + cytology)Specificity: 82% (ImmunoCyt),85% (cytology), NS(ImmunoCyt + cytology)Planz 2005 171Study design: C-SDTime period: NSCountry: GermanyEnrolled: 626; analysed: 495No previous history of BC:353; history of BC: 273Age (years): mean 62, rangeNSSex: NSTests and cut-off used:cytology (VU), subjectiveassessment; cytology (BW),subjective assessment;cytology (VU + BW)Unit of analysis: specimen(n = 346, cytology (VU);n = 191 cytology (BW);n = 535, cytology (VU) +cytology (BW))Sensitivity: 38% (cytology(VU)), 38% (cytology (BW)),39% (cytology (VU) +cytology (BW))Specificity: 98% (cytology(VU)), 99% (cytology (BW)),98% (cytology (VU) +cytology (BW))Ponsky 2001 139Study design: C-SDTime period: May 1996 toDec 1998Country: USAEnrolled: 608; analysed: 608No previous history of BC:529; history of BC: 79Age (years): mean 70(malignant group), 61 (benigngroup), range NSSex: 438 M, 170 FTests and cut-off used:NMP22 ≥ 10 U/ml; cytology(VU), subjective assessmentUnit of analysis: patient(n = 608)Sensitivity: 88% (NMP22),62% (cytology)Specificity: 84% (NMP22),85% (cytology)Potter 1999 172Study design: C-SDTime period: NSCountry: UKEnrolled: 336; analysed: 336No previous history of BC:336; history of BC: 0Age (years): mean 64, rangeNSSex: 336 MTests and cut-off used:cytology (VU), subjectiveassessmentUnit of analysis: patient(n = 336)Sensitivity: 100%Specificity: 99%Poulakis 2001 140aStudy design: C-SDTime period: NSCountry: Germany, USAEnrolled: 739; analysed: 739No previous history of BC:353; history of BC: 386Age (years): mean 67, range37 to 90Sex: 485 M, 254 FTests and cut-off used:NMP22 ≥ 8.25 U/ml; cytology(VU), subjective assessmentUnit of analysis: patient(n = 739)Sensitivity: 85% (NMP22),62% (cytology)Specificity: 68% (NMP22),96% (cytology)225© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Appendix 12Study Participants Tests Outcomes summaryRaitanen 2002 173,174Study design: CC-SD (18centres)Time period: 1997–9Country: FinlandEnrolled: 652; analysed: 570No previous history of BC:151; history of BC: 501Age (years): mean 69, range21 to 92Sex: 449 M, 121 FTests and cut-off used:cytology (VU), subjectiveassessmentUnit of analysis: patient(n = 129 no history of BC;n = 441, previous BC history)Sensitivity: 57% (no history),35% (BC history)Specificity: NS (no history),90% (BC history)Ramakumar 1999 159Study design: case–control(nhd)Time period: Sep 1997 toDec 1997Country: USAEnrolled: 196; analysed: 196NMP22, 112 cytologyNo previous history of BC:19; history of BC: 38Age (years): mean 66, range29 to 102Sex: 152 M, 44 FTests and cut-off used:NMP22 ≥ 10 U/ml; cytology(VU), subjective assessmentUnit of analysis: patient(n = 196, NMP22; n = 112,cytology)Sensitivity: 53% (NMP22),44% (cytology)Specificity: 60% (NMP22),95% (cytology)Saad 2002 141Study design: C-SDTime period: NSCountry: UKEnrolled: 120; analysed: 120No previous history of BC:120; history of BC: 0Age (years): mean 70, range30 to 88Sex: 100 M, 20 FTests and cut-off used:NMP22 ≥ 10 U/ml; cytology(VU), subjective assessmentUnit of analysis: patient(n = 120)Sensitivity: 81% (NMP22),48% (cytology)Specificity: 87% (NMP22),87% (cytology)Sanchez-Carbayo 1999 161a(primary report)Study design: case–control(nhd)Time period: NSCountry: SpainEnrolled: 267; analysed: 187No previous history of BC:NS; history of BC: NSAge (years): NSSex: NSTests and cut-off used:NMP22 ≥ 14.6 U/mlUnit of analysis: patient(n = 187)Sensitivity: 76%Specificity: 95%[Sanchez-Carbayo 1999 160 ](secondary report)Study design: case–control(nhd)Time period: NSCountry: SpainEnrolled: 267; analysed: 187No previous history of BC:NS; history of BC: NSAge (years): NSSex: NSTests and cut-off used:NMP22 ≥ 6.4, 7, 10, 12,13.7 U/mlUnit of analysis: patient(n = 187)Sensitivity: 81% (NMP2210 U/ml)Specificity: 91% (NMP2210 U/ml)Sanchez-Carbayo 2001 142Study design: C-SDTime period: NSCountry: SpainEnrolled: 232; analysed: 232No previous history of BC: 0;history of BC: 232Age (years): NSSex: 201 M, 31 FTests and cut-off used:NMP22 ≥ 10 U/mlUnit of analysis: patient(n = 232)Sensitivity: 69%Specificity: 93%Sanchez-Carbayo 2001 162Study design: case–control(nhd)Time period: Jan1999 to Jul1999Country: SpainEnrolled: 187; analysed: 187No previous history of BC:112; history of BC: 0Age (years): mean 66, range24 to 89Sex: 87 M, 25 FTests and cut-off used:NMP22 ≥ 10 U/ml; cytology(VU or catheterised)Unit of analysis: patient(n = 187, NMP22; n = 112,cytology)Sensitivity: 61% (NMP22),35% (cytology)Specificity: 80% (NMP22),97% (cytology)Sarosdy 2002 108Study design: case–control(nhd) (21 centres)Time period: NS to Apr 2000Country: USAEnrolled: 451; analysed: 392No previous history of BC: 0;history of BC: 176Age (years): mean 71 (cases),58 (control), range 25 to 98Sex: NSTests and cut-off used: FISH,aneuploidy of chromosomes3, 7 and 17 or loss of the9p21 locus; cytology (VU),subjective assessmentUnit of analysis: patient(n = 392, FISH)Sensitivity: 71% (FISH)Specificity: 84% (FISH)226

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Study Participants Tests Outcomes summarySarosdy 2006 103Study design: C-SD (23centres)Time period: NS to Apr 2003Country: USAEnrolled: 497; analysed: 473No previous history of BC:497; history of BC: 0Age (years): mean 63, range40 to 97Sex: 298 M, 199 FTests and cut-off used: FISH,NS; cytology (VU), subjectiveassessmentUnit of analysis: patient(n = 473)Sensitivity: 69% (FISH), 38%(cytology)Specificity: 78% (FISH), NS(cytology)Schmitz-Drager 2008 117,118Study design: CC-SDTime period: Oct 2000 toJul 2007Country: GermanyEnrolled: 301; analysed: 280No previous history of BC:301; history of BC: 0Age (years): mean 59(gross hematuria group),57 (microhematuria group),range 24 to 89Sex: 227 M, 65 FTests and cut-off used:ImmunoCyt, more than onegreen or red urothelial cell;cytology (VU), subjectiveassessment; cystoscopy, NS;ImmunoCyt + cystoscopy;cystoscopy + cytology (VU)Unit of analysis: patient(n = 280, ImmunoCyt; n = 280,cytology; n = 278, cystoscopy;n = 280, ImmunoCyt+cystoscopy; n = 280,cystoscopy + cytology)Sensitivity: 85% (ImmunoCyt),44% (cytology), 84%(cystoscopy), 100%(ImmunoCyt + cystoscopy),88% (cystoscopy + cytology)Specificity: 88% (ImmunoCyt),96% (cytology), 98%(cystoscopy), 87%(ImmunoCyt + cystoscopy),95% (cystoscopy + cytology)Serretta 2000 143,144Study design: C-SDTime period: NSCountry: ItalyEnrolled: 179; analysed: 179No previous history of BC: 0;history of BC: 179Age (years): mean 65, range31 to 84Sex: 151 M, 28 FTests and cut-off used:NMP22 ≥ 10 U/mlUnit of analysis: patient(n = 179)Sensitivity: 75%Specificity: 55%Shariat 2006 147Study design: C-SDTime period: NSCountry: AustriaEnrolled: 2951; analysed: 2871No previous history of BC: 0;history of BC: 2871Age (years): mean 68, range21 to 97Sex: 2166 M, 705 FTests and cut-off used:NMP22 ≥ 10 U/ml and1–30 U/mlUnit of analysis: patient(n = 2871)Sensitivity: 57% (NMP2210 U/ml)Specificity: 81% (NMP2210 U/ml)Sharma 1999 148Study design: C-SDTime period: NSCountry: USAEnrolled: 278; analysed: 278No previous history of BC:199; history of BC: 79Age (years): NSSex: NSTests and cut-off used:NMP22 ≥ 10 U/ml forpatients with no previoushistory of BC, ≥ 6 U/mlfor patients with previoushistory of BC; cytology (VU),subjective assessmentUnit of analysis: patient(n = 199, NMP22 10 U/ml;n = 278, cytology)Sensitivity: 67% (NMP2210 U/ml), 56% (cytology)Specificity: 86% (NMP2210 U/ml), 93% (cytology)Skacel 2003 104Study design: CC-SDTime period: 1996–2001Country: USAEnrolled: 120; analysed: 111No previous history of BC:26; history of BC: 94Age (years): NSSex: NSTests and cut-off used: FISH,chromosomal gain of twoor more chromosomes infive or more cells per slide,or in cases of isolated gainsof chromosome 3, 7 or 17when the number of cellswith such gain was ≥ 10%,or when 12 or more cellswith 9p21 loss was the onlyabnormalityUnit of analysis: patient(n = 111)Sensitivity: 85%Specificity: 97%227© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Appendix 12Study Participants Tests Outcomes summarySokolova 2000 105Study design: C-SDTime period: NSCountry: USAEnrolled: 179; analysed: 179No previous history of BC:86; history of BC: 93Age (years): NSSex: NSTests and cut-off used: FISH,five or more cells withpolysomy; cytology (VU),subjective assessmentUnit of analysis: patient(n = 179)Sensitivity: 85%Specificity: 92%Sozen 1999 163Study design: case–control(nhd)Time period: NSCountry: TurkeyEnrolled: 140; analysed: 140No previous history of BC:NS; history of BC: NSAge (years): mean 71 (cases),62 (controls), range NSSex: 127 M, 13 FTests and cut-off used:NMP22 ≥ 5, 6.4, 7, 10, 12,15 U/ml; cytology (VU orcatheterised), subjectiveassessmentUnit of analysis: patient(n = 140)Sensitivity: 73% (NMP2210 U/ml), 35% (cytology)Specificity: 81% (NMP2210 U/ml), 90% (cytology)Stampfer 1998 149Study design: C-SD (threecentres)Time period: NSCountry: USAEnrolled: 231; analysed: 217No previous history of BC: 0;history of BC: 231Age (years): mean 68, rangeNSSex: 166 M, 65 FTests and cut-off used:NMP22 ≥ 5, 6.4, 7, 10 U/ml;cytology (VU), subjectiveassessmentUnit of analysis: cystoscopy(n = 274, NMP22 10 U/ml;n = 200, cytology)Sensitivity: 49% (NMP2210 U/ml), 43% (cytology)Specificity: 92% (NMP2210 U/ml), 92% (cytology)Takeuchi 2004 164aStudy design: case–control(nhd)Time period: Nov 1999 toMay 2004Country: JapanEnrolled: 669; analysed: 669No previous history of BC:48; history of BC: 0Age (years): NSSex: NSTests and cut-off used:NMP22 ≥ 12 U/ml; cytology(VU), subjective assessment;NMP22 + cytology (VU)Unit of analysis: patient(n = 669, NMP22; n = 699,cytology; n = 48, NMP22 +cytology)Sensitivity: 58% (NMP22),44% (cytology), 60% (NMP22+ cytology)Specificity: 80% (NMP22),100% (Cytology), NS(NMP22 + cytology)Talwar 2007 150Study design: C-SDTime period: Mar 2004 toApr 2006Country: IndiaEnrolled: 196; analysed: 196No previous history of BC:127; history of BC: 63Age (years): mean 63, range39 to 78Sex: 142 M, 54 FTests and cut-off used:NMP22 ≥ 10 U/ml; cytology(VU), subjective assessmentUnit of analysis: patient(n = 196)Sensitivity: 67% (NMP22),22% (cytology)Specificity: 81% (NMP22),99% (cytology)Tetu 2005 119Study design: C-SDTime period: May 2000 toJul 2002Country: CanadaEnrolled: 904; analysed: 870No previous history of BC:NS; history of BC: NSAge (years): NSSex: NSTests and cut-off used:ImmunoCyt, one green orone red fluorescent cell;cytology (VU), subjectiveassessment; ImmunoCyt +cytology (VU)Unit of analysis: patient(n = 870)Sensitivity: 74% (ImmunoCyt),29% (Cytology), 84%(ImmunoCyt + cytology)Specificity: 62% (ImmunoCyt),98% (cytology), 61%(Immunocyt + cytology)Tritschler 2007 80Study design: C-SDTime period: Sep 2004 toApr 2005Country: GermanyEnrolled: 100; analysed: 100NMP22; 94 cytologyNo previous history of BC:30; history of BC: 70Age (years): mean 68, rangeNSSex: 71 M, 29 FTests and cut-off used:NMP22 ≥ 10 U/ml; cytology(VU), subjective assessment;cytology (BW), subjectiveassessmentUnit of analysis: patient(n = 100, NMP22; n = 85,cytology (VU); n = 94,cytology (BW))Sensitivity: 65% (NMP22),44% (cytology (VU)), 76%(cytology (BW))Specificity: 40% (NMP22),78% (cytology (VU)), 62%(cytology (BW))228

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Study Participants Tests Outcomes summaryWiener 1998 151Study design: C-SDTime period: Jan 1996 to Oct1996Country: AustriaEnrolled: 291; analysed: 291No previous history of BC:190; history of BC: 101Age (years): mean 62, range17 to 90Sex: 199 M, 92 FTests and cut-off used:NMP22 ≥ 10 U/ml; cytology(VU), subjective assessment;cytology (BW), subjectiveassessmentUnit of analysis: patient(n = 291, NMP22; n = 291,cytology (VU); n = 200,cytology (BW))Sensitivity: 48% (NMP22),59% (cytology (VU)), 58%(cytology (BW))Specificity: 69% (NMP22),100% (cytology (VU)), 100%(cytology (BW))Yoder 2007 106Study design: C-SDTime period: Jun 2002 toDec 2003Country: USAEnrolled: 250; analysed: 250No previous history of BC: 0;history of BC: 250Age (years): median 72, rangeNSSex: 187 M, 63 FTests and cut-off used: FISH,more than two chromosomalgains of chromosomes 3, 7 or17 in at least four analysedcells, or homozygous 9p21deletion in at least 12analysed cells, or isolatedtrisomy of chromosome 3,7 or 17 in at least 10% ofanalysed cellsUnit of analysis: patient(n = 250)Sensitivity: 64%Specificity: 73%Zippe 1999 152,153Study design: C-SDTime period: Apr 1997 toFeb 1998Country: USAEnrolled: 330; analysed: 330No previous history of BC:330; history of BC: 0Age (years): mean 63, rangeNSSex: 254 M, 76 FTests and cut-off used:NMP22 ≥ 10 U/ml; cytology(VU), subjective assessmentUnit of analysis: patient(n = 330)Sensitivity: 100% (NMP22)33% (cytology)Specificity: 86% (NMP22)100% (cytology)BW, bladder wash; C-SD, cross-sectional diagnostic study; CC-SD, consecutive cross-sectional diagnostic study; nhd, nocompletely healthy donors in control group; NS, not stated; VU, voided urine.a Studies used non-standard cut-off.229© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Appendix 13Quality assessment results for thebiomarker and cytology studies231© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Appendix 13232Study Marker Q1 Q2 Q3 Q4 Q5 Q6 Q7 Q8 Q9 Q10 Q11 Q12 Q13 Q14Abbate 1998 154 N + + + + ? + ? + ? + + + + –Bastacky 1999 165 C + + ? + + + ? ? ? + + + + –Bhuiyan 2003 120 N, C + + + + + + ? ? ? + + –N, ?C +N, –C –Boman 2002 155 N, C + + + + + + ? ? + + + + + –Casella 2000 121 N, a C + + + + + + ? ? ? + + + + –Casetta 2000 122 N, a C a + + + + + + ? ? ? + + +C, –N + –Chahal 2001 166 C a + + + + – + + ? ? + + + + –Chahal 2001 45 N, a C a + + + + ? + + + ? + + +C, –N + –Chang 2004 156 N + + ? ? ? + ? ? ? + + – + –Daniely 2007 94 F+C + + ? + + + + ? ? + + + + –Del Nero 1999 123 N, a C a + + + + + + + ? ? + + –N, ?C +N, –C –Friedrich 2003 95 N, a F a + + + + + + ? ? ? + + + + –Garbar 2007 167 C + + ? + + + + ? ? + + + + –Giannopoulos 2001 125 N + + + + + + + ? ? + + – + –[Giannopoulos 2000 124 ] N, C a + + + + + + + ? ? + + –N, +C + –Grossman 2005 126 N, a C a – + + + + + + + ? + + + + –Grossman 2006 127 N, a C a + + + + + + + + ? + + + + –Guttierez Banos 2001 128 N, a C a + + + + + + ? ? ? + + +N, ?C +N, –C –Hakenberg 2000 168 C + + ? + + + ? ? ? + + + – –Halling 2000 97 F, a C a + + + + + + + ? – + + + + –Hughes 1999 129 N, C + + – + + + ? ? ? + + + + +Junker 2006 98 F, a C + + + + + + ? ? ? + + + + –Karakiewicz 2006 170 C a + + + + + + ? ? – + + + + –Kipp 2008 99 F a + + – + + + + ? + + + + – –Kowalska 2005 130 N a + + + + + + ? ? ? + + ? – –+ + + + + + + + ? + + + + –Kumar 2006 131 N, a C, aN+CLahme 2001 132 N, a C a + + + + + + ? ? ? + + –N, ?C +N, –C –Lee 2001 157 N, C a + + + + + + ? + ? + + + + –Lodde 2003 109 I, a C, a I+C + + + + + + ? ? ? + + + + –Lodde 2006 110 I, C, I+C + + + + + + ? ? ? + + + + –May 2007 107 F, a C a + + ? + + + + ? + + + + – –

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Study Marker Q1 Q2 Q3 Q4 Q5 Q6 Q7 Q8 Q9 Q10 Q11 Q12 Q13 Q14Meiers 2007 100 F, a C a + + ? + + + ? ? ? + + + + –Messing 2005 111 I, a C, a I+C + + + + + + + ? ? + + + + +Mian 1999 112 I, a C, a I+C + + + + + + ? ? ? + + + + –Mian 2000 134 N a + + + + + + ? ? ? + + + + –+ + + + + + ? ? + + + + – –Mian 2003 101 F, a I, a C, aI+CMian 2006 113 I, C, I+C + + ? + – + ? ? ? + + + + –Miyanaga 1999 135 N, C a + + + + + + ? ? ? + + + + –Miyanaga 2003 136 N, C a + + + + + + ? + ? + + + + –Moonen 2007 102 F, C, F+C + + + + + + + ? + + + +F, ?C +F, –C –Oge 2001 137 N a + + + + + + ? ? ? + + + + –Olsson 2001 114 I, a C + + + + + + + ? ? + + +I, ?C +I, –C –Oosterhuis 2002 138 N + + + + + + ? ? ? + + – + –Parekattil 2003 158 N, C + + ? + + + + + ? +C, –N + +C, –N +C, –N –Piaton 2003 116 I, a C, a I+C + + + + + + +C, ?I ? ? + + + + –Planz 2005 171 C + + + + + + ? + ? + + + – –Ponsky 2001 139 N, a C a + + ? + ? ? ? ? ? + + +N, ?C +N, –C –Potter 1999 172 C a + + ? – – + + ? ? + + ? – –Poulakis 2001 140 N, a C a + + + + + + + ? ? + + +C, –N + –Raitanen 2002 174b C a + + + + + + + ? – + + + + –Ramakumar 1999 159 N, a C a + + + + + + ? + ? + + +C, –N + –Saad 2002 141 N, a C a + + + + + + ? ? ? + + +N, ?C +N, –C –Sanchez-Carbayo 1999 161 N, [N a ] + + + + + + ? + + + + ? + –(primary report)[Sanchez-Carbayo1999 160 ] (secondaryreport)Sanchez-Carbayo 2001 142 N a + + + + + + ? ? ? + + – + –Sanchez-Carbayo 2001 162 N, a C + + + + + + ? + ? + + –N, ?C +N, –C –Sarosdy 2002 108 F, a C + + + + + + + + ? + + +F, ?C +F, –C +C, –FSarosdy 2006 103 F, a C + + + + + + + + ? + + +F, ?C – –Schmitz–Drager 2008 118 I, a C a + + ? + ? + + ? ? + – + + –233© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Appendix 13234Study Marker Q1 Q2 Q3 Q4 Q5 Q6 Q7 Q8 Q9 Q10 Q11 Q12 Q13 Q14Serretta 2000 144 N a + + ? + + + ? ? ? + + + + –Shariat 2006 147 N a + + + + + + ? + ? + + + + –Sharma 1999 148 N, a C a + + ? + + + ? ? ? + + +N, ?C + –Skacel 2003 104 F a + + + + + + + + ? + + + + –Sokolova 2000 105 F, a C + + + + + + + + – + + –F, ?C +F, –C –Sozen 1999 163 N, a C + + ? ? ? + + + ? + + –N, ?C +N, –C –Stampfer 1998 149 N, C + + + + + + + ? ? ? + +C, –N + –Takeuchi 2004 164 N, C, a + + ? + + + ? ? ? + + + + –N+CTalwar 2007 150 N, a C a + + + + + + ? +C, ?N ? + + +N, ?C +N, –C –Tetu 2005 119 I, a C, a I+C + + + + + + ? ? ? + + + + –Tritschler 2007 80 N, a C a + + + + + + +N, ?C + ? + + + + –Wiener 1998 151 N, a C a + + + + + + ? ? ? + + + + –Yoder 2007 106 F a + + + + + + ? + ? + + + + –Zippe 1999 153 N, a C a + + ? + + + + ? ? + + + + –C, cytology; F, FISH; I, Immunocyt; N, NMP22; +, yes to the question; –, no to the question; ?, unclear.a Study included in the pooled estimates for this marker.b Although Raitanen 2002 174 did not report observer variation, Raitanen 2002 173 did.

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Appendix 14Studies of biomarkers included in pooledestimates for patient-level analysis and alsothose reporting specimen and stage/grade235© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Appendix 14236pTaG3–pT1 G1 G2 G1–2 G1 G3pTa,pT1G3pTa,pT1,CISpTa,pT1Study Marker Patient Specimen pTa pTaG1 pTaG1–2 pTaG2 pTaG3Abbate 1998 154 N üBastackyC ü1999 165Bhuiyan N, C ü2003 120Boman 2002 155 N, C ü ü üCasella 2000 121 N, a C ü ü üCasetta 2000 122bN, a C a üChahal 2001 166 C a ü ü ü üChahal 2001 45 N, a C a ü ü ü üChang 2004 156 N üDaniely 2007 94 F+C üDel Nero 1999 123N, a C a ü ü ü üFriedrich N, a F a ü ü ü ü2003 95[Friedrich N ü ü ü ü2002 96 ]Garbar 2007 167 C üGiannopoulos N ü ü ü ü2001 125C a ü ü ü ü[Giannopoulos2000 124 ]Grossman N, a C a ü ü ü ü ü2005 126Grossman N, a C a ü ü ü ü2006 127Guttierez N, a C a ü ü ü üBanos 2001 128Hakenberg C ü ü ü2000 168Halling 2000 97 F, a C a ü ü ü ü

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4pT3a,3b pT3G3 pT4pT2,pT2a pT2G2 pT2G3 pT2–3 pT2–4 ≥ pT2 pT3pT1G3 +CIS pT1–4 CIS G3 pT2Study pT1 pT1G1 pT1G2 pT1G3Abbate 1998 154Bastacky1999 165Bhuiyan2003 120Boman 2002 155 üCasella 2000 121 üCasetta2000 122bChahal 2001 166 ü ü ü ü ü ü üChahal 2001 45 ü ü ü üChang 2004 156Daniely 2007 94Del Nero ü ü1999 123Friedrich ü ü ü ü2003 95[Friedrich ü ü ü ü2002 96 ]Garbar 2007 167 üGiannopoulos ü ü ü ü2001 125ü ü ü ü[Giannopoulos2000 124 ]Grossman ü ü ü ü ü ü2005 126Grossman ü ü ü ü ü ü2006 127Guttierez ü ü üBanos 2001 128Hakenbergü ü2000 168Halling 2000 97 ü ü ü237© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Appendix 14238pTaG3–pT1 G1 G2 G1–2 G1 G3pTa,pT1G3pTa,pT1,CISpTa,pT1Study Marker Patient Specimen pTa pTaG1 pTaG1–2 pTaG2 pTaG3Hughes N, C ü1999 129N, C ü[Hutterer2008 169 ]Junker 2006 98 F, a C üKarakiewicz C a ü ü ü2006 170Kipp 2008 99 F a ü ü ü üKowalska N a ü2005 130Kumar 2006 131 N, a C, a ü ü ü ü ü üN+CLahme 2001 132 N, a C a ü ü ü ü üLee 2001 157 N, C a ü ü ü üLodde 2003 109 I, a C, a I+C ü ü ü üLodde 2006 110c I, C, I+C üMay 2007 107 F, a C a ü ü ü üMeiers 2007 100 F, a C a ü ü üMessing 2005 111I, a C, a I+C ü ü ü üMian 1999 112 üI, a C, a I+C ü ü ü üMian 2000 134 N a ü ü ü üMian 2003 101 F, a I, a C, a ü ü ü üI+CMian 2006 113 I, C, I+C ü ü ü üMiyanaga N, C a 1999 135Miyanaga N, C a ü2003 136Moonen F, C, F+C ü ü ü ü2007 102Oge 2001 137 N a ü ü ü üOlsson 2001 114 I, a C ü

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4pT2,pT2a pT2G2 pT2G3 pT2–3 pT2–4 ≥ pT2 pT3pT1G3 +CIS pT1–4 CIS G3 pT2Study pT1 pT1G1 pT1G2 pT1G3Hughes 1999 129[Hutterer2008 169 ]Junker 2006 98Karakiewiczü ü2006 170Kipp 2008 99 ü ü üKowalska2005 130Kumar 2006 131 ü ü üLahme 2001 132 ü ü üLee 2001 157 ü ü üLodde 2003 109 ü ü ü üLodde 2006 110cMay 2007 107 ü ü ü üMeiers 2007 100 üMessing 2005 111ü ü ü üMian 1999 112 ü ü ü üMian 2000 134 ü ü ü üMian 2003 101 ü ü ü üMian 2006 113 ü ü ü üMiyanaga1999 135pT3a,3b pT3G3 pT4Miyanaga2003 136Moonen ü ü ü2007 102Oge 2001 137 ü ü üOlsson 2001 114239© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Appendix 14240pTaG3–pT1 G1 G2 G1–2 G1 G3pTa,pT1G3pTa,pT1,CISpTa,pT1Study Marker Patient Specimen pTa pTaG1 pTaG1–2 pTaG2 pTaG3Oosterhuis N ü ü2002 138Parekattil N, C ü2003 158d Piaton 2003116I, a C, a I+C ü ü ü ü üPlanz 2005 171 C ü ü üPonsky 2001 139 N, a C a üPotter 1999 172 C a üPoulakis 2001 140N, a C a ü ü ü ü üd Raitanen C a ü ü ü ü2002 174d Ramakumar N, a C a ü ü ü ü1999 159Saad 2002 141 N, a C a ü ü ü üSanchez-Carbayo1999 161 N ü ü ü üN a ü[Sanchez-Carbayo1999 160 ]Sanchez- N a ü ü ü2001 142CarbayoSanchez- N, a C ü ü ü ü2001 162eCarbayoSarosdy F, a C ü ü ü ü ü2002 108Sarosdy F, a C ü ü ü ü ü ü2006 103Schmitz- I, a C a üDrager 2008 118SerrettaN a ü ü ü ü2000 144

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4pT3a,3b pT3G3 pT4pT2,pT2a pT2G2 pT2G3 pT2–3 pT2–4 ≥ pT2 pT3pT1G3 +CIS pT1–4 CIS G3 pT2Study pT1 pT1G1 pT1G2 pT1G3Oosterhuis ü ü ü2002 138Parekattil2003 158d Piaton 2003116ü ü üPlanz 2005 171 üPonsky 2001 139Potter 1999 172Poulakis ü ü ü ü ü ü ü2001 140d Raitanen ü ü ü ü2002 174d Ramakumarü ü1999 159Saad 2002 141 ü ü ü üSanchez- Carbayo1999 161ü ü ü ü ü üü ü[Sanchez-Carbayo1999 160 ]Sanchez-ü ü ü ü2001 142CarbayoSanchez-Carbayo2001 162e ü ü ü ü ü üSarosdyü ü2002 108Sarosdy ü ü ü2006 103Schmitz-üDrager 2008 118Serretta ü ü ü2000 144241© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Appendix 14242pTaG3–pT1 G1 G2 G1–2 G1 G3pTa,pT1G3pTa,pT1,CISpTa,pT1Study Marker Patient Specimen pTa pTaG1 pTaG1–2 pTaG2 pTaG3Shariat 2006 147 N a üSharma N, a C a ü1999 148Skacel 2003 104f F a ü ü ü üSokolova F, a C ü ü ü2000 105Sozen 1999 163 ü ü ü ü üN, a C üStampfer N, C 1998 149gd Takeuchi N, C, a N+C ü ü ü2004 164Talwar 2007 150 üN, a C a ü ü üTetu 2005 119 I, a C, a I+C ü üTritschler N, a C a ü ü 2007 80hWiener N, a C a ü ü ü ü1998 151Yoder 2007 106 F a üZippe 1999 153 N, a C a ü

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4pT3a,3b pT3G3 pT4pT2,pT2a pT2G2 pT2G3 pT2–3 pT2–4 ≥ pT2 pT3pT1G3 +CIS pT1–4 CIS G3 pT2Study pT1 pT1G1 pT1G2 pT1G3Shariat 2006 147Sharma 1999 148 ü ü ü üSkacel 2003 104f ü ü ü üSokolova2000 105Sozen 1999 163 ü ü ü ü üStampfer ü ü1998 149gd Takeuchiü ü ü2004 164Talwar 2007 150 ü ü üTetu 2005 119 ü ü ü üTritschler 2007 80hü ü üWiener ü1998 151Yoder 2007 106Zippe 1999 153a Included in the meta-analysis models for that biomarker.b Casetta 2000 122 – NMP22 (cut-off 10 U/ml) 2 × 2 data reported only for the subgroup of patients with a history of bladder cancer.c Lodde 2006 110 – unit of analysis was tumour.d Stage/grade categories not on grid because of insufficient space: pT1G1–2; 116 ≥ pTa + CIS; 116 G2–3; 174 pT1–T3b; 159 CIS–pT1. 164e Sanchez-Carbayo 2001 162 – stage and grade information reported only for NMP22 (not cytology).f Skacel 2003 104 – stage and grade information reported only for FISH (not cytology).g Stampfer 1998 149 – stage and grade information reported only for NMP22 at cut-off of ≥ 6.4 U/ml (not cytology) with cystoscopy (not patient or specimen) as the unit of analysis.h Tritschler 2007 80 – stage and grade information reported only for NMP22 at cut-off 10 U/ml and bladder wash cytology (not voided urine cytology), with tumour as the unit of analysis(not patient).243© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Appendix 15Biomarker and cytology testperformance for detecting bladdercancer, results table with 2 × 2 data245© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Appendix 15246Specificity(%)Sensitivity(%)Numberanalysed TP FP FN TNStudy Test Cut-off Unit of analysis12 U/ml Patient 135 59 4 50 22 54 85NMP22 (Matritechtest kit)Laboratory analysisAbbate 1998 154No. of patients 109,of whom no previoushistory of BC NS, historyof BC NS, plus benigngenitourinary disorders 26Patient 743 115 39 65 524 64 93Cytology (VU orBW)Bastacky 1999 165No. of patients 1672, ofwhom no previous historyof BC 752, history of BC485, other 435≥ 10 U/ml Specimen 25 94NMP22 (Matritech)Laboratory analysisBhuiyan 2003 120NMP22 (voided) 3.6 U/ml Specimen 231 42 66 27 96 61 59NMP223.6 U/ml Specimen 54 61(cystoscopicallycollected)Cytology (VU) Specimen 125 27 3 40 55 40 95Cytology(cystoscopicallycollected)Specimen 62 83No. of patients 233, ofwhom no previous historyof BC NS, history of BC NS≥ 4 U/ml Specimen? 297 83 46 71 97 54 68No previous history of BC 127 42 16 23 46 65 74pTaG1–2 39 23 16 59pTaG3–pT1 20 13 7 65≥ pT2 6 6 0 100NMP22 (Matritech)Laboratory analysisBoman 2002 155No. of patients 250, ofwhom no previous historyof BC NS, history of BC174≤ 10 mm 17 11 6 6511–20 mm 24 13 11 5421–30 mm 15 11 4 73> 30 mm 9 8 1 89

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Specificity(%)Sensitivity(%)Numberanalysed TP FP FN TNStudy Test Cut-off Unit of analysisPrevious history of BC 170 41 30 48 51 46 63pTaG1–2 68 25 43 37pTaG3–pT1 13 10 3 77≥ pT2 8 6 2 75≤ 10 mm 73 30 43 4111–20 mm 12 8 4 67> 21 mm 5 3 2 60Cytology (BW) Specimen? 293 60 12 90 131 40 93No previous history of BC 125 26 4 37 58 41 94pTaG1–2 38 8 30 21pTaG3–pT1 19 13 6 68≥ pT2 6 5 1 83≤ 10 mm 17 6 11 3511–20 mm 24 8 16 3321–30 mm 9 5 4 55> 30 mm 8 7 1 87Previous history of BC 168 34 8 53 73 39 92pTaG1–2 67 20 47 30pTaG3–pT1 12 9 3 75≥ pT2 8 5 3 63≤ 10 mm 74 22 52 3011–20 mm 11 10 1 91> 21 mm 5 5 0 100247© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Appendix 15248Specificity(%)Sensitivity(%)Numberanalysed TP FP FN TNStudy Test Cut-off Unit of analysisNMP22 (Matritech) 10 U/ml Patient 235 67 17 63 88 52 84Laboratory analysisSuperficial77 28 49 36(non-invasive)G1 42 13 29 31G2 50 28 22 56G3 38 26 12 68Cytology (BW) Patient 200 50 11 45 94 53 90Superficial65 25 40 38(non-invasive)G1 30 9 21 30G2 38 19 19 50G3 24 22 2 92Casella 2000 121No. of patients 235, ofwhom no previous historyof BC NS, history of BC NS≥ 10 U/ml Patient 229 85 25 37 82 70 77NMP22 (Matritech)Laboratory analysisPatient 191 50 10 43 88 54 90Cytology (BW) Only highgradeatypiawas consideredpositive[Shariat 2003 145 ]No. of patients 229, ofwhom no previous historyof BC NS, history of BC NSNMP22 ≥ 10 U/ml Patient 209 58 14 59 78 50 85pTa 65 23 42 35pT1 31 23 8 74≥ pT2 17 12 5 71CIS 4 0 4 0G1 41 12 29 29G2 41 22 19 54G3 35 24 11 69[Shariat 2004 146 ]No. of patients 209, ofwhom no previous historyof BC NS, history of BCNS, controls 92

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Specificity(%)Sensitivity(%)Numberanalysed TP FP FN TNStudy Test Cut-off Unit of analysisPatient 174 46 8 43 77 52 91pTa 65 23 42 35pT1 31 25 6 81≥ pT2 17 16 1 92CIS 4 3 1 75G1 41 12 29 29G2 41 18 23 45G3 35 33 2 95Cytology (BW) Only highgradeatypiawas consideredpositiveNMP22 11 U/ml Patient 196 84 20 66 26 56 57NMP22 12 U/ml No previous history 94 35 8 45 6 44 43of BCNMP22 10 U/ml History of BC 102 45 12 25 20 64 63Cytology (VU) Dubious cases Patient 196 110 9 40 37 73 80consideredNo previous positiveof BChistory 94 61 1 19 13 76 93History of BC 102 49 8 21 24 70 75Cytology (VU) Dubious cases Patient 196 88 5 62 41 59 89consideredNo previous negativeof BChistory 94 48 1 32 13 60 93History of BC 102 40 4 30 28 58 88Casetta 2000 122No. of patients 196, ofwhom no previous historyof BC 94, history of BC 102Patient 285 23 14 24 224 49 94pTaG1 11 3 8 27pTaG2 12 3 9 25pTaG3 3 2 1 67pT1G1 1 0 1 0pT1G2 5 4 1 80pT1G3 6 4 2 67pT1G3 and CIS 1 1 0 100pT2G2 1 1 0 100pT2G3 5 3 2 60pT3G3 2 2 0 100Cytology (VU) Suspicious classedwith positiveChahal 2001 166No. of patients 285, ofwhom no previous historyof BC NS, history of BC NS249© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Appendix 15250Specificity(%)Sensitivity(%)Numberanalysed TP FP FN TNStudy Test Cut-off Unit of analysis10 U/ml Patient 211 11 14 22 164 33 92pTa 22 2 20 9pT1 7 3 4 43pT2 3 1 2 33pT3 1 1 0 100G1 19 1 18 5G2 6 2 4 33G3 8 4 4 50No previous history of 96 7 7 9 73 44 91BCPrevious history of BC 115 4 8 13 90 24 92Patient 211 8 5 25 173 24 97pTa 22 4 18 18pT1 7 5 2 72pT2 3 1 2 33pT3 1 1 0 100G1 19 3 16 16G2 6 2 4 33G3 8 6 2 67No previous history of 96 5 4 11 76 31 95BCPrevious history of BC 115 2 1 15 97 13 99NMP22 (test kit)Laboratory analysisChahal 2001 45No. of patients 211, ofwhom no previous historyof BC 96, history of BC 115Cytology (VU) (thesepatients alreadyincluded for cytologyin Chahal 2001 166 ?)≥ 7.5 U/ml Patient 314 4 52 7 251 36 83NMP22 (Sancordon)Laboratory analysisChang 2004 156No. of patients 399, ofwhom benign urothelialdisease or urogenitalcancer 331 (11 BC), noprevious history of BC NS,history of BC NS

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Specificity(%)Sensitivity(%)Numberanalysed TP FP FN TNStudy Test Cut-off Unit of analysisPatient 115 12 51 0 52 100 50FISH: minimumof four cellswith gains oftwo or morechromosomes or≥ 12 cells withhomozygous lossof the 9p21 locusFISH (UroVysion) +cytology (VU)Daniely 2007 94No. of patients 115, ofwhom no previous historyof BC 49, history of BC 66NMP2210 U/ml Patient 105 62 4 13 26 83 87Laboratory analysispTa 30 20 10 67pT1 45 42 3 93G1 29 20 9 69G2 36 34 2 94G3 10 8 2 806 U/ml Patient 105 69 7 6 23 92 76pTa 30 25 5 83pT1 45 44 1 98G1 29 25 4 86G2 36 35 1 97G3 10 9 1 905 U/ml Patient 105 70 11 5 19 94 63Cytology (VU) Patient 105 35 5 40 25 47 83pTa 30 6 24 20pT1 45 29 16 64G1 29 11 18 38G2 36 16 20 44G3 10 8 2 80Del Nero 1999 123No. of patients 105, ofwhom no previous historyof BC 0, history of BC 105251© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Appendix 15252Specificity(%)Sensitivity(%)Numberanalysed TP FP FN TNStudy Test Cut-off Unit of analysis≥ 10 U/ml Patient 103 32 20 14 37 70 65pTa 21 11 10 52pT1 18 17 1 94≥ pT2 6 5 1 83CIS 1 1 0 100G1 7 3 4 43G2 28 20 8 70G3 11 9 2 83Patient 103 31 6 15 51 67 89pTa 21 13 8 62pT1 18 12 6 67≥ pT2 6 5 1 83CIS 1 1 0 100G1 7 4 3 57G2 28 18 10 63G3 11 9 2 83NMP22 (Matritech)Laboratory analysisFriedrich 2003 95No. of patients 103, ofwhom no previous historyof BC 55, history of BC 48FISH (UroVysion) If 20% of thecells had a gainof two or morechromosomes(3, 7 or 17) or40% of the cellshad a gain of onechromosome or40% loss of 9p21locusNMP22 10 U/ml Specimen 146 37 32 17 60 69 65pTa 25 13 12 52pT1 20 18 2 90≥ pT2 8 6 2 75CIS 1 0 1 0G1 7 3 4 43G2 31 22 9 71G3 16 12 4 75[Friedrich 2002 96 ]No. of patients 115, ofwhom no previous historyof BC 70, history of BC 45Specimen 592 98 22 65 407 60 95CIS 50 38 12 76Low-grade atypia 59 22 37 37High-grade atypia 14 3 11 21Cytology (BW) Suspicious classedas positiveGarbar 2007 167No. of patients 139, ofwhom no previous historyof BC NS (82 BW), historyof BC NS (510 BW)

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Specificity(%)Sensitivity(%)Numberanalysed TP FP FN TNStudy Test Cut-off Unit of analysis213 75 27 43 68 64 72> 8 U/ml Patient (excluding healthyvolunteers)NMP22 (Matritech)Laboratory analysispTa 57 30 27 53pT1 32 21 11 66pT2–4 20 18 2 90CIS 6 5 1 83G1 30 15 15 50G2 45 25 20 56G3 43 35 8 81No previous history of 68 50 18 74BCHistory of BC 50 28 22 56Healthy volunteers 21 2 19 90Urological disease 50 16 34 68No evidence of disease 45 11 34 76Excluding stones, urinary tract infection andurological malignanciesother than BC211 75 18 43 75 64 81Giannopoulos 2001 125No. of patients 234, ofwhom no previous historyof BC 118, history of BC95, healthy volunteers 21NMP22 > 8 U/ml Patient 168 62 18 37 51 63 74pTa 49 26 23 53pT1 26 17 9 65pT2–T4 16 14 2 88CIS 5 4 1 80G1 26 14 12 54G2 39 21 18 54G3 34 27 7 79No evidence of disease 25 7 18 72Urological disease 23 9 14 72[Giannopoulos 2000 124 ]No. of patients 168, ofwhom no previous historyof BC 85, history of BC 62,healthy volunteers 21253© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Appendix 15254Specificity(%)Sensitivity(%)Numberanalysed TP FP FN TNStudy Test Cut-off Unit of analysis147 38 4 61 44 38 92Cytology (VU) Patient (excluding 21healthy volunteers)pTa 49 8 41 16pT1 26 12 14 46pT2–T4 16 11 5 69CIS 5 5 0 100G1 26 2 24 8G2 39 10 29 26G3 34 26 8 77No evidence of disease 25 4 21 84Urological disease 9 0 9 10010 U/ml Patient 1331 44 179 35 1073 56 86pTa 30 14 16 47pT1 27 13 14 48pTa, T1 62 31 31 50pT2, T2a 6 6 0 100pT3a, T3b 4 3 1 75pT2–T3 10 9 1 90CIS 5 4 1 80pTx 7 4 3 57G1 27 13 14 48G2 18 9 9 50G3 25 18 7 72Gx (grade unknown) 9 4 5 44No urinary tract disease 567 55 512 90Benign prostatic 280 49 231 83hypertrophy/prostatitisCystitis/inflammation/ 125 28 97 78trigonitis/urinary tractinfectionErythema 51 9 42 82NMP22 (Matritech)BladderChekGrossman 2005 126No. of patients 1331, ofwhom no previous historyof BC 1331, history of BC 0

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Specificity(%)Sensitivity(%)Numberanalysed TP FP FN TNStudy Test Cut-off Unit of analysis53 12 41 77Hyperplasia/squamousmetaplasia/cysts andpolypsCalculi 40 11 29 73Trabeculations 217 42 175 81Other benign diseases, 220 41 179 81kidney and genitourinaryOther cancer history, 8 1 7 88non-bladderOther active cancer, nonbladder38 5 33 87Cytology (VU) Patient 1287 12 10 64 1201 16 99pTa 28 2 26 7pT1 27 5 22 19pTa, T1 60 10 50 17pT2, T2a 6 2 4 33pT3a, T3b 3 0 3 0pT2–T3 9 2 7 22CIS 5 3 2 60pTx 7 0 7 0G1 25 0 25 0G2 18 3 15 17G3 24 9 15 38Gx (grade unknown) 9 0 9 0Cystoscopy (NS) Patient 79 70 9 89NMP22 + cystoscopy Patient 79 74 5 94255© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Appendix 15256Specificity(%)Sensitivity(%)Numberanalysed TP FP FN TNStudy Test Cut-off Unit of analysis≥ 10 U/ml Patient 668 51 72 52 493 50 87pTa 50 18 32 36pT1 17 11 6 65pT2 8 7 1 88pT3 1 1 0 100pT4 2 2 0 100CIS 8 4 4 50pTx 17 8 9 47pTa, pT1, CIS 75 33 42 44pTa, pT1, CIS, G1 53 19 34 36pT2–T4 11 10 1 91G1 38 12 26 32G2 16 7 9 44G3 32 24 8 75All grades 86 43 43 50Poorly differentiated 33 24 9 73muscle-invasive G3, T2–T4No evidence of urinary 264 28 236 89tract diseaseBenign prostatic 120 17 103 86hyperplasia/prostatitisErythema/cystitis/ 82 12 70 85inflammationUrinary tract infection 3 3 0 0Hyperplasia/squamous 23 2 21 91metaplasia/cyst/polyp/carbuncleCalculi 6 1 5 83Trabeculations 49 6 43 88Diverticulum/pouch/ cellule18 3 15 83NMP22 (Matritech)BladderChekGrossman 2006 127No. of patients 668, ofwhom no previous historyof BC 0, history of BC 668

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Specificity(%)Sensitivity(%)Numberanalysed TP FP FN TNStudy Test Cut-off Unit of analysisCytology (VU) Patient 650 12 17 86 535 12 97pTa 48 3 45 6pT1 16 2 14 13pT2 7 0 7 0pT3 1 0 1 0pT4 2 0 2 0CIS 8 3 5 38pTx 16 4 12 25pTa, pT1, CIS 72 8 64 11pTa, pT1, CIS, G1 50 2 48 4pT2–T4 10 0 10 0G1 37 2 35 5G2 14 0 14 0G3 31 6 25 19All grades 82 8 74 10Poorly differentiated 32 6 26 19muscle-invasive G3, T2–T4Cystoscopy (NS) Patient 103 94 9 91G3 32 24 8 75NMP22 + cystoscopy Patient 103 102 1 99G3 32 31 1 97Cytology +cystoscopyPatient 103 97 6 94257© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Appendix 15258Specificity(%)Sensitivity(%)Numberanalysed TP FP FN TNStudy Test Cut-off Unit of analysisNMP2210 U/ml Patient 150 58 7 18 67 76 91Laboratory analysispTa 16 8 8 50pT1 46 39 7 80pT2–T4 14 13 1 93G1 16 8 8 50G2 29 20 9 69G3 31 30 1 97NMP226 U/ml Patient 150 64 10 12 64 84 87Laboratory analysispTa 16 11 5 69pT1 46 39 7 85pT2–T4 14 14 0 100G1 16 11 5 69G2 29 22 7 76G3 31 31 0 100Cytology (VU) Patient 150 53 5 23 69 70 93pTa 16 6 10 38pT1 46 35 1 76pT2–T4 14 12 2 86G1 16 7 9 44G2 29 18 11 62G3 31 28 3 90Cystoscopy (R) Patient 150 76 8 0 66 100 89Gutierrez Banos 2001 128No. of patients 150, ofwhom no previous historyof BC 64, history of BC 86

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Specificity(%)Sensitivity(%)Numberanalysed TP FP FN TNStudy Test Cut-off Unit of analysisCytology (VU) Specimen 417 257 16 80 64 76 80CIS 16 13 3 81G1 57 27 30 47G2 166 130 36 78G3 98 87 11 89Undergoing primary TUR 326 213 11 65 37 77 77CIS 11 8 3 73G1 48 26 22 54G2 136 106 30 78G3 83 73 10 88Undergoing secondary 91 44 5 15 27 75 84TURCIS 5 5 0 100G1 9 1 8 11G2 30 24 6 80G3 15 14 1 93Hakenberg 2000 168No. of patients 374, ofwhom no previous historyof BC 374, history of BC 0Patient 151 59 3 14 75 81 96pTa 37 24 13 65pT1–T4 19 18 1 95CIS 17 17 0 100G1 11 4 7 36G2 25 19 6 76G3 37 36 1 97Patient 118 40 1 29 48 58 98pTa 36 17 19 47pT1–T4 15 9 6 60CIS 18 14 4 78G1 11 3 8 27G2 24 13 11 54G3 34 24 10 71FISH (UroVysion) Five or more cellswith polysomyHalling 2000 97No. of patients 265, ofwhom no previous historyof BC 115, previous historyof BC 150Cytology (VU) Suspicious classedwith positive259© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Appendix 15260Specificity(%)Sensitivity(%)Numberanalysed TP FP FN TNStudy Test Cut-off Unit of analysis≥ 6.4 U/ml Specimen 128 25 16 28 59 47 79NMP22 (Matritech)Laboratory analysisHughes 1999 129Specimen 128 32 31 21 44 60 58Cytology (VU) Indeterminateresults wereclassed as negativeNo. of patients 107, ofwhom no previous historyof BC 0, history of BC 107Patient 121 57 5 38 21 60 81FISH (UroVysion) Five or more cellsshowed gains ofmore than onechromosome (3,7 or 17), or ≥10 cells showedgains of a singlechromosome (3,7 or 17), or ≥10 cells showedhomozygous lossof 9p21 locusJunker 2006 98No. of patients 141, ofwhom no previous historyof BC NS, history of BC NSCytology (NS) Patient 109 21 2 66 20 24 91Cytology (VU) Patient 2542 404 87 494 1557 45 95pTa–T1, CIS 2542 284 92 408 1758 41 95≥ pT2 2542 122 117 84 2219 59 95G1–G2 2542 211 97 393 1841 35 95G3 2542 197 112 97 2136 67 95Karakiewicz 2006 170No. of patients 2686, ofwhom no previous historyof BC 0, history of BC 26861731 62 301 18 1350 78 8210 U/ml Patients with non-TCCrecurrenceNMP22Laboratory analysis1731 16 86 64 1565 20 95Cytology (VU) Patients with non-TCCrecurrence[Hutterer 2008 169 ]No. of patients 1731, ofwhom no previous historyof BC 0, history of BC 1731

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Specificity(%)Sensitivity(%)Numberanalysed TP FP FN TNStudy Test Cut-off Unit of analysisPatient 124 53 5 32 34 62 87pTaG1 22 7 15 32pTaG2 11 7 4 64pTaG3 4 3 1 75pT1 10 8 2 80≥ pT2 16 14 2 88CIS 21 13 8 62Small cell carcinoma 1 1 0 100With muscle-invasive BC 17 15 2 88FISH (UroVysion) Four or more cellshad polysomicsignal patterns(gain of two ormore of the fourchromosomesin an individualcell), ≥ 10 cellsdemonstratedtetrasomy (foursignal patterns forall four probes) or> 20% of the cellsdemonstrated9p21 homozygousdeletion (loss ofthe two 9p21signals)Kipp 2008 99No. of patients 124, ofwhom no previous historyof BC 41, history of BC81, previous cancer of theupper urinary tract 2Patient 124 57 6 28 33 67 85pTaG1 22 21 1 96pTaG2 11 9 2 82pTaG3 4 3 1 75pT1 10 10 0 100≥ pT2 16 13 3 81CIS 21 16 5 76Small cell carcinoma 1 1 0 100With muscle-invasive BC 17 10 7 59Patient 124 74 8 11 31 87 79With muscle-invasive BC 17 16 1 94Cystoscopy (NS) Suspicious classedwith positiveFISH (UroVysion) +cystoscopy10 U/ml? Patient 98 10 43 9 36 53 46NMP22Laboratory analysisKowalska 2005 130No. of patients 98, of whomno previous history of BC0, history of BC 98261© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Appendix 15262Specificity(%)Sensitivity(%)Numberanalysed TP FP FN TNStudy Test Cut-off Unit of analysisNMP22 (Matritech) 10 U/ml Patient 131 39 19 7 66 85 78BladderChekpTa 21 16 5 76pT1 17 15 2 88≥ pT2 8 8 0 100G1 11 9 2 82G2 22 18 4 81G3 13 12 1 92Low risk (pTaG1–G2) 18 15 3 83High risk (pTaG3, T1) 20 16 4 80≥ invasive pT2 8 8 0 100Cytology (VU) Patient 131 19 3 27 82 41 96pTa 21 3 18 14pT1 17 8 9 47≥ pT2 8 8 0 100G1 11 2 9 19G2 22 6 16 27G3 13 11 2 85Low risk (pTaG1–G2) 18 2 16 11High risk (pTaG3, T1) 20 9 11 45≥ invasive pT2 8 8 0 100NMP22 +cytology (VU)Patient 46 42 4 91Kumar 2006 131No. of patients 131, ofwhom no previous historyof BC 0, history of BC 131

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Specificity(%)Sensitivity(%)Numberanalysed TP FP FN TNStudy Test Cut-off Unit of analysis109 25 27 15 42 63 6110 U/ml Patient (excluding healthycontrols)NMP22 (Matritech)Laboratory analysisLahme 2001 13284 25 15 15 29 63 66With BC or beingfollowed upNo. of patients 169, ofwhom no previous historyof BC 40, history of BC 44,other 25, healthy controls60Ta 22 9 13 41T1 8 5 3 63Ta-T1 30 14 16 47T2–4 10 9 1 90G1 25G2 68G3 100Healthy controls 60 4 56 93With benign lesions 25 12 13 51Cytology (VU) Patient (excluding healthy 109 18 5 22 64 45 93controls)With BC or being 84 18 5 22 39 45 89followed upTa 22 6 16 27T1 8 5 3 63Ta-T1 30 11 19 37T2–4 10 7 3 70G1 20G2 59G3 67Healthy controls 60 0 60 100With benign lesions 25 0 25 100263© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Appendix 15264Specificity(%)Sensitivity(%)Numberanalysed TP FP FN TNStudy Test Cut-off Unit of analysisNMP22 (Matritech) 7.7 U/ml Patient 106 53 10 17 26 76 72Laboratory analysispTa 23 13 10 57pT1 23 18 5 78p ≥ T2 24 22 2 92G1 19 11 8 58G2 34 27 7 79G3 17 15 2 88Cytology (VU) Patient 106 39 4 31 32 56 89pTa 23 8 15 35pT1 23 13 10 57p ≥ T2 24 18 6 75G1 19 4 15 21G2 34 22 12 65G3 17 13 4 77Lee 2001 157No. of patients 106, ofwhom no previous historyof BC NS, history of BC NS(cases 70; controls 36 withhistory of BC)Patient 225 89 40 13 83 87 67pTa 62 50 12 81pT1 16 15 1 94≥ pT2 11 10 1 91CIS 13 13 0 100G1 43 35 8 81G2 28 25 3 89G3 31 29 2 94No previous history of BC 91 47 10 4 30 92 75pTa 29 25 4 86pT1 13 13 0 100≥ pT2 6 5 1 83CIS 3 3 0 100G1 20 17 3 85G2 18 18 0 100G3 13 12 1 92ImmunoCyt At least onegreen or one redfluorescent cellLodde 2003 109No. of patients 235, ofwhom no previous historyof BC 98, history of BC 137

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Specificity(%)Sensitivity(%)Numberanalysed TP FP FN TNStudy Test Cut-off Unit of analysisHistory of BC 134 42 30 9 53 82 64pTa 33 25 8 76pT1 3 2 1 67≥ pT2 5 5 0 100CIS 10 10 0 100G1 23 18 5 78G2 10 7 3 70G3 18 17 1 94Cytology (VU) Patient 225 42 7 60 116 41 94pTa 62 7 55 11pT1 16 14 2 88≥ pT2 11 10 1 91CIS 13 12 1 92G1 43 2 41 5G2 28 12 16 43G3 31 28 3 90No previous history of BC 91 22 2 29 38 43 95pTa 29 4 25 14pT1 13 11 2 85≥ pT2 6 5 1 83CIS 3 3 0 100G1 20 1 19 5G2 18 10 8 56G3 13 11 2 85History of BC 134 20 5 31 78 39 94pTa 33 3 30 9pT1 3 3 0 100≥ pT2 5 5 0 100CIS 10 9 1 90G1 23 1 22 4G2 10 2 8 20G3 18 17 1 94265© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Appendix 15266Specificity(%)Sensitivity(%)Numberanalysed TP FP FN TNStudy Test Cut-off Unit of analysisPatient 225 92 40 10 83 90 68pTa 62 52 10 84pT1 16 16 0 100≥ pT2 11 11 0 100CIS 13 13 0 100G1 43 35 8 81G2 28 27 1 55G3 31 31 0 100No previous history of BC 91 48 10 3 94 75pTa 29 25 4 86pT1 13 13 0 100≥ pT2 6 6 0 100CIS 3 3 0 100G1 20 17 3 85G2 18 18 0 100G3 13 13 0 100History of BC 134 44 30 7 53 86 64pTa 33 27 6 82pT1 3 3 0 100≥ pT2 5 5 0 100CIS 10 10 0 100G1 23 18 5 78G2 10 9 1 90G3 18 18 0 100ImmunoCyt +cytology (VU)

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Specificity(%)Sensitivity(%)Numberanalysed TP FP FN TNStudy Test Cut-off Unit of analysisImmunoCyt At least one Specimen 334 85 51 16 182 71 78green or one redfluorescent cellLow-risk group 132 26 21 4 81 87 79Intermediate-risk group 124 35 19 8 62 81 77High-risk group 78 24 11 4 39 86 78Cytology (VU) Specimen 277 49 9 52 167 49 95Low-risk group 132 5 1 25 101 17 99Intermediate-risk group 67 20 4 23 20 47 83High-risk group 78 24 4 4 46 86 92ImmunoCyt +Specimen 334 87 52 14 181 86 78cytology (VU)Low-risk group 132 26 21 4 81 87 79Intermediate-risk group 124 36 20 7 61 84 75High-risk group 78 25 11 3 39 89 78Lodde 2006 110No. of patients 216, ofwhom no previous historyof BC 0, history of BC 216Patient 166 33 27 29 77 53 74pTa 38 16 22 42pT1 10 8 2 80pT2 12 8 4 67CIS 2 1 1 50G1 33 14 19 42G2 14 5 9 36G3 15 14 1 93No previous history of 58 32 26 55BC groupHistory of BC group 4 1 3 25FISH (UroVysion) Gain of twoor morechromosomesin five or morecells per slide,or in cases ofisolated gains ofchromosome 3, 7,or 17 when theproportion of cellswith such a gainwas 10% or moreof at least 100cells evaluated, orwhen there were≥ 10 cells with9p21 lossMay 2007 107No. of patients 166, ofwhom no previous historyof BC 62, history of BC 71and other 33267© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Appendix 15268Specificity(%)Sensitivity(%)Numberanalysed TP FP FN TNStudy Test Cut-off Unit of analysisCytology (VU) Patient 166 44 17 18 87 71 84pTa 38 16 22 58pT1 10 10 0 100pT2 12 10 2 83CIS 2 2 0 100G1 33 17 16 52G2 14 13 1 93G3 15 14 1 93No previous history of 58 41 17 71BC groupHistory of BC group 4 3 1 75Patient 624 158 44 12 410 93 90G1 42 36 6 42G2 48 45 3 94G3 80 77 3 96FISH (UroVysion) Chromosomalgain of twoor morechromosomes (+3,+7, +17) in fouror more cells ordeletion of 9p21in 12 or morecellsMeiers 2007 100No. of patients 624, ofwhom no previous historyof BC NS, history of BC NSCytology (VU) Patient 624 124 59 46 395 73 87G1 42 21 21 50G2 48 38 10 79G3 80 65 15 81

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Specificity(%)Sensitivity(%)Numberanalysed TP FP FN TNStudy Test Cut-off Unit of analysisImmunoCyt One confirmed Patient 326 42 68 10 206 81 75red or greenfluorescent cellpTa 35 29 6 83pT1 8 6 2 75pT2 2 2 0 100CIS 5 5 0 100G1 28 22 6 79G2 10 9 1 90G3 6 4 2 67Tumour< 1 cm 17 12 5 711–3 cm 19 16 3 84> 3 cm 5 3 2 60Cytology (VU) Patient 326 12 19 40 255 23 93pTa 35 2 33 6pT1 8 4 4 50pT2 2 0 2 0CIS 5 5 0 100G1 28 2 26 7G2 10 1 9 10G3 6 4 2 67Tumour< 1 cm 17 3 14 181–3 cm 19 5 14 26> 3 cm 5 1 4 20ImmunoCyt +Patient 326 42 73 10 201 81 73cytology (VU)pTa 35 29 6 83pT1 8 6 2 75pT2 2 2 0 100CIS 5 5 0 100G1 28 22 6 79G2 10 9 1 90G3 6 4 2 67Messing 2005 111No. of patients 341, ofwhom no previous historyof BC 0, history of BC 341269© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Appendix 15270Specificity(%)Sensitivity(%)Numberanalysed TP FP FN TNStudy Test Cut-off Unit of analysisImmunoCyt At least one Patient 249 68 35 11 135 86 79green or one redfluorescent cellpTa 43 37 6 86pT1 20 17 3 85≥ pT2 12 10 2 83CIS 4 4 0 100G1 25 21 4 84G2 25 21 4 84G3 29 26 3 90Cytology (VU) Patient 249 37 3 42 167 47 98pTa 43 9 34 21pT1 20 14 6 70≥ pT2 12 10 2 83CIS 4 4 0 100G1 25 1 24 4G2 25 13 12 52G3 29 23 6 79ImmunoCyt +Patient 249 71 35 8 135 90 79cytology (VU)pTa 43 38 5 88pT1 20 18 2 90≥ pT2 12 11 1 92CIS 4 4 0 100G1 25 21 4 84G2 25 22 3 88G3 29 28 1 97Mian 1999 112No. of patients 264, ofwhom no previous historyof BC 114, history of BC150

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Specificity(%)Sensitivity(%)Numberanalysed TP FP FN TNStudy Test Cut-off Unit of analysis≥ 10 U/ml Patient 240 30 39 24 147 56 79pTa 29 15 14 52pT1 13 6 7 46≥ pT2 10 7 3 70CIS 2 2 0 100G1 18 9 9 50G2 20 10 10 50G3 16 11 5 69NMP22 (Matritech)Laboratory analysisMian 2000 134No. of patients 240, ofwhom no previous historyof BC 81, history of BC 159Patient 181 69 29 11 72 86 71pTa 47 38 9 81pT1 13 12 1 92≥ pT2 10 9 1 90CIS 10 10 0 100G1 31 25 6 81G2 24 21 3 88G3 25 23 2 92Patient 57 27 16 1 13 96 45pTa 26 25 1 96pT1 1 1 0 100CIS 1 1 0 100G1 8 7 1 88G2 19 19 0 100G3 1 1 0 100ImmunoCyt At least onegreen or one redfluorescent cellMian 2003 101No. of patients 181, ofwhom no previous historyof BC 81, history of BC 100FISH (UroVysion) Four or moreaneusomic of 25counted cells271© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Appendix 15272Specificity(%)Sensitivity(%)Numberanalysed TP FP FN TNStudy Test Cut-off Unit of analysisCytology (VU) Patient 181 36 6 44 95 45 94pTa 47 6 41 13pT1 13 12 1 92≥ pT2 10 9 1 90CIS 10 9 1 90G1 31 2 29 6G2 24 11 13 46G3 25 23 2 92ImmunoCyt +Patient 181 72 34 8 67 90 66cytology (VU)pTa 47 39 8 83pT1 13 13 0 100≥ pT2 10 10 0 100CIS 10 10 0 100G1 31 26 5 84G2 24 22 2 92G3 25 24 1 96ImmunoCyt At least one Specimen 1886 253 436 45 1152 85 73green or one redfluorescent cellpTa 202 165 37 82pT1 47 42 5 89≥ pT2 19 16 3 84CIS 28 28 0 100G1 121 96 25 79G2 88 74 14 84G3 89 82 7 92Cytology (VU) Specimen 1886 116 10 182 1578 39 99pTa 202 41 161 20pT1 47 31 16 68≥ pT2 19 17 2 90CIS 28 26 2 93G1 121 10 111 8Mian 2006 113No. of patients 942, ofwhom no previous historyof BC 0, history of BC 942

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Specificity(%)Sensitivity(%)Numberanalysed TP FP FN TNStudy Test Cut-off Unit of analysisG2 88 39 49 43G3 89 67 22 75Specimen 1886 266 436 32 1152 89 73pTa 202 172 30 85pT1 47 46 1 98≥ pT2 19 17 2 90CIS 28 28 0 100G1 121 96 25 79G2 88 80 8 91G3 89 88 1 99ImmunoCyt +cytology (VU)309 20 68 2 219 91 7612 U/ml Patient (urothelial cancer,not just BC)NMP22 (Konica-Matritech)Laboratory analysisMiyanaga 1999 135309 12 1 10 286 55 100Cytology (VU) Patient (urothelial cancer,not just BC)No. of patients 309, ofwhom no previous historyof BC 309, history of BC 0NMP22 (Matritech) ≥ 12 U/ml Patient 137 8 14 35 80 19 85Laboratory analysis≥ 5 U/ml Patient 137 21 32 22 62 49 66Cytology (VU) Patient 137 3 2 40 92 7 98Miyanaga 2003 136No. of patients 156, ofwhom no previous historyof BC 99, history of BC 57Specimen 103 25 4 39 35 39 90pTa 44 12 32 27pT1 10 6 4 60pT2–4 4 2 2 50G1 27 6 21 22G2 19 7 12 37G3 18 12 6 67FISH (UroVysion) Four or moreof the 25morphologicallyabnormal cellsshowed gainsof two or morechromosomes (3,7 or 17) or ≥ 12of the 25 cells hadno 9p21 signalsMoonen 2007 102No. of patients 105, ofwhom no previous historyof BC 0, history of BC 105273© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Appendix 15274Specificity(%)Sensitivity(%)Numberanalysed TP FP FN TNStudy Test Cut-off Unit of analysisCytology (VU) Specimen 108 26 4 38 40 41 90pTa 44 12 32 27pT1 10 7 3 70pT2–4 4 3 1 75G1 27 6 21 22G2 19 5 14 26G3 18 15 3 83FISH (UroVysion) +Specimen 103 34 8 30 31 53 79cytology (VU)pTa 44 18 26 40pT1 10 8 2 80pT2–4 4 3 1 75G1 27 8 19 30G2 19 10 9 53G3 18 16 2 8976 37 8 13 18 74 69Patient (excluding thosewith benign urologicaldisease and healthyvolunteers)10 U/ml (studyalso reportssensitivity andspecificity at 5 and20 U/ml)NMP22 (Matritech)Laboratory analysisOge 2001 137No. of patients 114, ofwhom no previous historyof BC 37, history of BC 39,benign urological conditions18, healthy subjects 20pTaG1 12 5 7 42pTaG2 8 6 2 75pT1G2 6 5 1 83G1 G3 2 2 0 100pT2G2 4 4 0 100CIS G3 18 15 3 83Patient 114 31 26 0 57 100 69ImmunoCyt At least one greenor one red cellOlsson 2001 114Cytology (BW) Patient 114 18 13 58No. of patients 121, ofwhom no previous historyof BC 60, history of BC 61

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Specificity(%)Sensitivity(%)Numberanalysed TP FP FN TNStudy Test Cut-off Unit of analysis10 U/ml Specimen 431 32 116 32 251 50 68pTa 39 13 26 33Low grade 32 7 25 22High grade 7 6 1 86pT1 17 14 3 82Low grade 1 0 1 0High grade 16 14 2 88≥ pT2 2 2 0 100CIS 4 1 3 25NMP22 (Matritech)Laboratory analysisOosterhuis 2002 138No. of patients 191, ofwhom no previous historyof BC 0, history of BC 1912.5 U/ml Patient 252 19 123 8 102 70 45NMP22Laboratory analysisParekattil 2003 158Patient 253 18 43 9 183 67 81Atypia classedwith positiveCytology (VU orBW)No. of patients 253, ofwhom no previous historyof BC 155, history of BC 98Patient 651 106 89 40 416 73 82With no previous history 215 44 26 15 130 75 83of BCpTaG1–G2 22 15 7 68pTa–1 G3 14 11 3 79pT1G1–G2 4 3 1 75≥ pT2 11 8 3 73CIS 1 1 0 100≥ pTa + CIS 5 4 1 80pTxGx (bladder) 1 1 0 100pTxGx (UUT) 1 1 0 100G1 10 4 6 40G2 17 15 2 88G3 30 23 7 77Gx 2 2 0 100With previous history of BC436 62 63 25 286 71 82ImmunoCyt At least onegreen or one redfluorescent cellPiaton 2003 116No. of patients 694, ofwhom no previous historyof BC 236, history of BC458275© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Appendix 15276Specificity(%)Sensitivity(%)Numberanalysed TP FP FN TNStudy Test Cut-off Unit of analysispTaG1–G2 42 27 15 64pTa–1 G3 18 16 2 89pT1G1–G2 3 2 1 67≥ pT2 16 10 6 63CIS 2 1 1 50≥ pTa + CIS 3 3 0 100pTxGx (bladder) 1 1 0 100pTxGx (UUT) 2 2 0 100G1 21 13 8 62G2 24 16 8 67G3 39 30 9 77Gx 3 3 0 100Cytology (VU) Patient 651 90 74 56 431 62 85With no previous history 215 42 26 17 130 71 83of BCpTaG1–G2 22 10 12 45pTa–1 G3 14 13 1 93pT1G1–G2 4 4 0 100≥ pT2 11 9 2 82CIS 1 0 1 0≥ pTa + CIS 5 4 1 80pTxGx (bladder) 1 1 0 100pTxGx (UUT) 1 1 0 100G1 10 3 7 30G2 17 12 5 71G3 30 25 5 83Gx 2 2 0 100With previous history of BC436 48 48 39 301 55 86

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Sensitivity(%)Numberanalysed TP FP FN TNStudy Test Cut-off Unit of analysisSpecificity(%)pTaG1–G2 42 19 23 45pTa–1 G3 18 14 4 78pT1G1–G2 3 3 0 100≥ pT2 16 9 7 56CIS 2 2 0 100≥ pTa + CIS 3 0 3 0pTxGx (bladder) 1 N/SpTxGx (UUT) 2 1 1 50G1 21 8 13 38G2 24 14 10 58G3 39 25 14 64Gx 3 1 2 33Patient 146 120 26 82With no previous history 59 51 8 86of BCpTaG1–G2 22 16 6 73pTa–1 G3 14 14 0 100ImmunoCyt +cytology (VU)pT1G1–G2 4 4 0 100≥ pT2 11 9 2 82CIS 1 1 0 100≥ pTa + CIS 5 5 0 100pTxGx (bladder) 1 1 0 100pTxGx (UUT) 1 1 0 100G1 10 5 5 50G2 17 16 1 94G3 30 28 2 93Gx 2 2 0 100With previous history of 87 69 18 79BCpTaG1–G2 42 30 10 71pTa–1 G3 18 16 2 89277© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Appendix 15278Specificity(%)Sensitivity(%)Numberanalysed TP FP FN TNStudy Test Cut-off Unit of analysispT1G1–G2 3 3 0 100≥ pT2 16 13 3 81CIS 2 2 0 100≥ pTa + CIS 3 3 0 100pTxGx (bladder) 1 1 0 100pTxGx (UUT) 2 2 0 100G1 21 13 8 62G2 24 19 5 79G3 39 34 5 87Gx 3 3 0 100ImmunoCyt At least one Patient 691 102 87 41 461 71 84green or one redfluorescent cellpTa 75 52 23 75pT1 28 19 9 67≥ pT2 28 20 8 73CIS 8 8 0 100G1 31 19 12 61G2 40 30 10 76G3 68 52 16 77No previous history of 58 42 16 72BCHistory of BC 85 60 25 71Cystoscopy Patient 691 141 79 2 469 99 86No previous history of BC58 58 0 100[Pfister 2003 115 ]No. of patients 694, ofwhom no previous historyof BC 236, history of BC458

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Specificity(%)Sensitivity(%)Numberanalysed TP FP FN TNStudy Test Cut-off Unit of analysisCytology (VU) Patient 691 70 30 73 518 49 95pTa 75 27 48 36pT1 28 17 11 59≥ pT2 28 25 3 91CIS 8 3 5 33G1 31 6 25 18G2 40 19 21 46G3 68 43 25 64No previous history of 58 37 21 64BCHistory of BC 85 33 52 39ImmunoCyt +Patient 691 113 106 30 442 79 81cytology (VU)pTa 75 56 19 75pT1 28 22 6 78≥ pT2 28 25 3 91CIS 8 8 0 100G1 31 21 10 67G2 40 31 9 78G3 68 59 9 87No previous history of 58 49 9 84BCHistory of BC 85 64 21 75Cytology (VU) Specimen 346 54 4 88 200 38 98One specimen 142 25 1 32 84 44 99Two specimens 142 32 1 25 84 56 99Three specimens 142 38 3 44 82 67 97Planz 2005 171No. of patients 626, ofwhom no previous historyof BC 353, history of BC273279© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Appendix 15280Specificity(%)Sensitivity(%)Numberanalysed TP FP FN TNStudy Test Cut-off Unit of analysisCytology (BW) Specimen 191 16 2 26 147 38 99One specimen 20 4 0 6 10 40 100Two specimens 20 5 0 5 10 50 100Three specimens 20 6 0 4 10 60 100Cytology (VU + BW) Specimen 535 70 6 112 347 39 98G1 42 5 37 12One specimen 13 2 11 15Two specimens 13 2 11 15Three specimens 13 2 11 15G2 55 23 32 42One specimen 27 13 14 48Two specimens 27 16 11 59Three specimens 27 20 7 74G3 45 26 19 58One specimen 23 12 11 52Two specimens 23 15 8 65Three specimens 23 18 5 78> 10 U/ml Patient 608 46 89 6 467 88 84NMP22Laboratory analysisPonsky 2001 139Cytology (VU) Patient 608 32 85 20 471 62 85No. of patients 608, ofwhom no previous historyof BC 529, history of BC 79Cytology (VU) Patient 336 2 3 0 331 100 99Potter 1999 172No. of patients 336, ofwhom no previous historyof BC 336, history of BC 0

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Specificity(%)Sensitivity(%)Numberanalysed TP FP FN TNStudy Test Cut-off Unit of analysisNMP22 (Matritech) ≥ 8.25 U/ml Patient 739 347 107 59 226 85 68Laboratory analysis (study also gives sensitivity andpTa 210 174 36 83specificity from pT1 47 40 7 856.4 to 20 U/ml) pT2 58 55 3 95pT3 45 43 2 96pT4 9 9 0 100CIS 31 23 8 74Superficial invasive (pTa, 286 237 49 83pT1, CIS)Superficial invasive (pT2– 111 107 4 96T4)G1 129 106 23 82G2 167 149 18 89G3 70 66 4 94With one tumour 208 165 43 79With two to three 92 83 9 90tumoursWith more than three 99 96 3 97tumoursWith no history of BC 179 154 25 86With history of BC 220 190 30 86≥ 10 U/ml Patient 739 321 101 85 232 79 70Cytology (VU) Patient 739 253 14 153 319 62 96pTa 211 93 118 44pT1 47 33 14 70pT2 58 45 13 78pT3 45 42 3 93pT4 9 8 1 89CIS 31 26 5 84Superficial invasive (pTa, pT1, CIS)287 152 135 53Poulakis 2001 140No. of patients 739, ofwhom no previous historyof BC 353, history of BC386281© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Appendix 15Specificity(%)282Sensitivity(%)Numberanalysed TP FP FN TNStudy Test Cut-off Unit of analysis112 95 17 85Superficial invasive (pT2–T4)G1 129 49 80 38G2 160 109 51 68G3 70 63 7 90With one tumour 208 99 109 48With two to three 92 63 29 68tumoursWith more than three 99 85 14 86tumoursWith no history of BC 179 140 39 78With history of BC 220 107 113 49PatientCytology (VU) PapanicolaouI–II classed asnegative and III–Vclassed as positive(suspicious classedas positive)Raitanen 2002 173129 73 56 57With no previous historyof BCNo. of patients 652, ofwhom no previous historyof BC 151, history of BC501pTa 51 16 35 35pT1 49 33 16 67pT2–T4 22 19 3 86pTx 3 3 0 100CIS (no suspicious) 4 2 2 50G1 44 10 34 23G2 45 28 17 62G3 39 34 5 87Gx (only suspicious) 1 1 0 100One tumour 86 49 37 57Two tumours 26 14 12 54More than threetumours14 10 4 71

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Specificity(%)Sensitivity(%)Numberanalysed TP FP FN TNStudy Test Cut-off Unit of analysis441 41 32 77 291 35 90With previous history ofBCpTa 54 9 45 16pT1 20 6 14 30> pT2 3 2 1 67pTx 33 13 20 39CIS 8 6 2 75G1 48 10 38 21G2 35 16 19 46G3 (no suspicious) 3 3 0 100Gx 32 12 20 38PatientWith no previous history 129 40 89 31of BCpTa 51 6 45 12pT1 49 19 30 39pT2-T4 22 11 11 50pTx 3 2 1 67CIS (no suspicious) 4 2 2 50G1 44 5 39 11G2 45 12 33 27G3 39 23 16 59Gx (only suspicious) 1 0 1 0One tumour 86 27 59 31Two tumours 26 5 21 19More than three14 8 6 57tumoursPapanicolaouI–III classed asnegative and IV–Vclassed as positive(suspicious classedas negative)283© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Appendix 15284Specificity(%)Sensitivity(%)Numberanalysed TP FP FN TNStudy Test Cut-off Unit of analysis441 21 11 97 312 18 97With previous history ofBCpTa 54 8 46 15pT1 20 4 16 20> pT2 3 0 3 0pTx 33 5 28 15CIS 8 4 4 50G1 48 6 42 13G2 35 7 28 20G3 (no suspicious) 3 3 0 100Gx 32 5 27 16Patient 575 71 8 177 319 29 98With no previous history 129 50 79 39of BCpTa–T1 100 32 68 32G1 44 3 41 7G2–3 84 47 37 56With history of BC 446 21 8 98 319 18 98Patient 575 61 11 187 316 25 97With no previous history 129 40 89 31of BCpTa–T1 100 25 75 25G1 44 5 39 11G2–3 84 35 49 42With history of BC 446 21 11 98 316 18 97[Raitanen 2002 174 ] Cytology (VU) PapanicolaouI–III classed asnegative and IV–Vclassed as positive(suspicious classedas negative) (localanalysis)PapanicolaouI–III classed asnegative and IV–Vclassed as positive(suspicious classedas negative)(review analysis)

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Specificity(%)Sensitivity(%)Numberanalysed TP FP FN TNStudy Test Cut-off Unit of analysisNMP22 (Matritech) 10 U/ml Patient 196 30 56 27 83 53 60Laboratory analysispTa 25 12 13 48pT1–T3b 14 11 3 79CIS 11 5 6 45G1 9 4 5 44G2 26 16 10 62G3 13 8 5 62No previous history of 19 14 5 74BC tumourHistory of BC tumour 38 16 22 42Cytology (VU) Patient 112 24 3 30 55 44 95pTa 24 7 17 29pT1–T3b 12 8 4 67CIS 11 8 3 73G1 9 2 7 22G2 24 9 15 38G3 12 10 2 83No previous history of 17 10 7 59BC tumourHistory of BC tumour 37 14 23 38Ramakumar 1999 159No. of patients 196, ofwhom no previous historyof BC 19, history of BC38 and control (othersundergoing cystoscopywho were negative for BCand with negative urinespecimen) 139≥ 10 U/ml Patient 120 42 9 10 59 81 87pTa 23 16 7 70pT1 20 18 2 90pT2 8 8 0 100CIS 6 6 0 100G1 13 8 5 62G2 22 19 3 86G3 17 15 2 88NMP22Laboratory analysisSaad 2002 141No. of patients 120, ofwhom no previous historyof BC 120, history of BC 0285© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Appendix 15286Specificity(%)Sensitivity(%)Numberanalysed TP FP FN TNStudy Test Cut-off Unit of analysisCytology (VU) Patient 120 25 9 27 59 48 87pTa 23 7 16 30pT1 20 10 10 50pT2 8 7 1 88CIS 6 5 1 83G1 13 2 11 15G2 22 9 13 41G3 17 14 3 82187 84 4 27 72 76 95> 14.6 U/ml Patient (from group 1: 111with positive cystoscopyfor TCC of the bladder;from group 2: 76 withprevious BC and free ofdisease at time of study)NMP22 (Matritech)Laboratory analysisSanchez-Carbayo 1999 160No. of patients 267, ofwhom no previous historyof BC NS, history ofBC NS, other urologicaldiseases 25, malignanciesof non-bladder origin 25,healthy volunteers 30Patient (from group 1)pTa 24 15 9 63pT1 53 42 11 79pT2 11 9 2 82pT3 13 13 0 100pT4 3 3 0 100CIS 3 2 1 67G1 33 25 8 76G2 33 27 6 82G3 40 33 7 83Single tumour 25 18 7 72Multiple tumours 81 61 20 75< 0.5 cm 23 19 4 830.5–3 cm 42 34 8 81> 3 cm 41 38 3 93Papillary tumour 78 63 15 81Solid tumour 28 28 0 100No previous history of 43 33 10 77BCPrevious history of BC 64 54 10 84

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Specificity(%)Sensitivity(%)Numberanalysed TP FP FN TNStudy Test Cut-off Unit of analysis6.4 U/ml Patient 187 95 15 16 61 85 807 U/ml Patient 187 95 13 16 63 85 8310 U/ml Patient 187 90 7 21 69 81 9112 U/ml Patient 187 87 6 24 70 78 9213.7 U/ml Patient 187 87 3 24 73 78 96NMP22 (Matritech)Laboratory analysis[Sanchez-Carbayo 1999 161 ]No. of patients as forSanchez-Carbayo 1999 160≥ 10 U/ml Patient 232 38 13 17 164 69 93Not receiving intravesical 106 25 5 6 70 81 93instillationsReceiving intravesical 126 13 8 11 94 65 92instillationsCIS 4 1 3 25G1 29 8 21 27G2 37 21 16 58G3 40 30 10 76NMP22 (Matritech)Laboratory analysisSanchez-Carbayo 2001 162No. of patients 232, ofwhom no previous historyof BC 0, history of BC 232≥ 10 U/ml Patient 187 26 29 17 115 61 80Patients with no previous 112 26 7 17 62 60 90history of BCpTa 5 0 5 0pT1 28 16 12 57CIS 1 0 1 0pT2 7 6 1 86pT3 2 2 0 100G1 11 2 9 18G2 15 9 6 60G3 17 14 3 82Patients with no previous 112 15 23 28 46 35 97history of BCNMP22 (Matritech)Laboratory analysisSanchez-Carbayo 2001 142No. of patients 187, ofwhom no previous historyof BC 112, history of BC 0,other 75Cytology (VU orcatheterised)287© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Appendix 15288Specificity(%)Sensitivity(%)Numberanalysed TP FP FN TNStudy Test Cut-off Unit of analysis392 44 54 18 276 71 84Patient (whole groupexcluding 59 healthydonors)FISH (UroVysion) Aneuploidy ofchromosomes 3, 7and 17 or loss ofthe 9p21 locusSarosdy 2002 108176 44 39 18 75 71 66Patients with history ofBCNo. of patients 451, ofwhom no previous historyof BC 0, history of BC 176and other 275pTaG1–G2 26 16 10 62pTaG3 6 5 1 83pT1G2 2 2 0 100pT1G3 3 3 1 75pT2 7 7 0 100CIS 7 7 0 100G1 22 12 10 55G2 9 7 2 78G3 18 17 1 94Control group 275 15 260 95Healthy donors 59 0 59 100Non-genitourinary 48 4 44 92benign diseaseNon-genitourinary 3 1 2 67cancerBenign prostatic58 5 53 92hyperplasiaMicrohaematuria 15 2 13 87Infections, inflammation 28 1 27 96Prostate and renal cancer61 5 56 92

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Specificity(%)Sensitivity(%)Numberanalysed TP FP FN TNStudy Test Cut-off Unit of analysisCytology (VU) PatientpTaG1–G2 26 6 20 23pTaG3 6 2 4 33pT1G2 2 2 0 100pT1G3 4 2 2 50pT2 3 1 2 33CIS 6 2 4 33G1 22 4 18 18G2 9 4 5 44G3 17 7 10 41FISH (UroVysion) NS (assay was Patient 473 35 94 16 328 69 78performedaccording topTaG1 21 10 11 48instructions on pTaG2 6 5 1 83the product pTaG3 4 4 0 100labelling)pT1 7 6 1 86pT2 10 9 1 90Unknown stage 2 1 1 50G1 21 10 11 48G2 10 7 3 70G3 17 15 2 88Cytology (VU) Patient 473 19 32 38pTaG1 21 5 16 24pTaG2 6 3 3 50pTaG3 4 2 2 50pT1 7 3 4 43pT2 10 6 4 60Unknown stage 2 1 1 50G1 21 5 16 24G2 10 3 7 30G3 17 9 8 53Sarosdy 2006 105No. of patients 497, ofwhom no previous historyof BC 497, history of BC 0289© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Appendix 15290Specificity(%)Sensitivity(%)Numberanalysed TP FP FN TNStudy Test Cut-off Unit of analysisImmunoCyt More than one Patient 280 23 31 4 222 85 88green or one redurothelial cellGross haematuria 63 15 8 2 38 88 83High grade 11 9 2 82Low grade 5 5 0 100Microhaematuria 217 8 23 2 184 80 89High grade 4 4 0 100Low grade 6 4 2 67Cytology (VU) Patient 280 12 11 15 242 44 96Gross haematuria 63 8 4 9 42 47 91High grade 11 6 5 55Low grade 5 2 3 40Microhaematuria 217 4 7 6 200 40 97High grade 4 2 2 50Low grade 6 4 2 67Cystoscopy (NS) Patient 278 21 4 4 249 84 98Gross haematuria 61 13 2 2 44 87 96High grade 11 8 3 73Low grade 5 4 1 80Microhaematuria 217 8 2 2 205 80 99ImmunoCyt +Patient 280 27 32 0 221 100 87cystoscopyGross haematuria 63 17 9 0 37 100 80Microhaematuria 217 10 23 0 184 100 89Cystoscopy +Patient 280 22 13 5 240 88 95cytology (VU)Gross haematuria 63 14 6 3 40 82 87Microhaematuria 217 8 7 2 200 80 97Schmitz-Drager 2008 117,118No. of patients 301, ofwhom no previous historyof BC 301, history of BC 0

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Specificity(%)Sensitivity(%)Numberanalysed TP FP FN TNStudy Test Cut-off Unit of analysis10 U/ml Patient 179 41 56 14 68 75 55pTa 13 7 6 54pT1 27 20 7 74pT2–T3 12 10 2 83CIS 3 3 0 100G1 7 3 4 43G2 19 12 7 63G3 29 26 3 90NMP22 (Matritech)Laboratory analysisSerretta 2000 144No. of patients 179, ofwhom no previous historyof BC 0, history of BC 179NMP22 ≥ 10 U/ml Patient 137 30 37 12 58 71 6120 U/ml Patient 137 24 18 18 77 57 81[Serretta 1998 143 ]No. of patients 137, ofwhom no previous historyof BC 0, history of BC 137Patient 2871 596 347 449 1479 57 81≥ pT2 220 183 37 83G3 329 247 82 75≥ 10 U/ml(study also givessensitivity andspecificity forcut-offs from 1 to30 U/ml)NMP22Laboratory analysisShariat 2006 147No. of patients 2871, ofwhom no previous historyof BC 0, history of BC 2871Patient 278 28 44 6 200 82 82≥ 10 U/ml forpatients with noprevious historyof BC, ≥ 6 U/mlfor patients withhistory of BCNMP22 (Matritech)Laboratory analysisSharma 1999 148No. of patients 278, ofwhom no previous historyof BC 199, history of BC 79199 4 27 2 166 67 86≥ 10 U/ml With no previous historyof BC79 24 17 4 34 86 67≥ 6 U/ml With previous history ofBCPatient 278 19 17 15 227 56 93With no previous history 199 2 8 4 185 33 96of BCWith previous history of BC79 17 9 11 42 61 82Cytology (VU) Atypical classedwith positive291© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Appendix 15292Specificity(%)Sensitivity(%)Numberanalysed TP FP FN TNStudy Test Cut-off Unit of analysisPatient 278 10 1 24 243 29 100With no previous history 199 1 0 5 193 17 100of BCWith previous history of BC79 9 1 19 50 32 98Cytology (VU) Atypical classedwith negativePatient 111 70 1 12 28 85 97pTa 64 53 11 83pT1 6 5 1 83pT2 6 6 0 100pT4 3 3 0 100CIS 3 3 0 100G1 23 19 4 83G2 35 28 7 80G3 24 23 1 96FISH (UroVysion) Chromosomalgain of twoor morechromosomesin five or morecells per slide,or in cases ofisolated gain ofchromosome 3, 7or 17 when thenumber of cellswith such gain was≥ 10%, or when9p21 loss was theonly abnormality,≥ 12 cells withsuch lossSkacel 2003 104No. of patients 120, ofwhom no previous historyof BC 26, history of BC 94179 39 11 7 122 85 92Patient (urothelialcarcinoma)FISH (UroVysion) Five or more cellswith polysomySokolova 2000 105pTa 22 14 8 65pT1–T4 12 11 1 95CIS 12 12 0 100G2 14 11 3 76G3 22 21 1 97CEP3 19 14 5 74No. of patients 179, ofwhom no previous historyof BC 86, history of BC 932.8% of cells withtetrasomyCEP7 21 16 5 766.5% of cells withtetrasomy

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Specificity(%)Sensitivity(%)Numberanalysed TP FP FN TNStudy Test Cut-off Unit of analysisCEP8 19 11 8 58CEP9 21 11 10 52CEP11 19 10 9 53CEP17 21 13 8 627.1% of cells withtetrasomy6.2% of cells withtetrasomyCEP18 19 8 11 427.0% of cells withtetrasomy9p21 21 6 15 2916.9% of cellswith homozygousdeletion179 41 16 5 117 90 88Patient (urothelialcarcinoma)Four or more cellswith polysomyCEP3, 7, 17 and 9p21 21 20 1 95Cytology (VU) Patient (urothelialcarcinoma)pTa 22 10 12 47pT1–T4 12 7 5 60CIS 12 9 3 78G2 14 8 6 54G3 22 16 6 715 U/ml Patient 140 36 37 4 63 90 636.4 U/ml Patient 140 35 31 5 69 88 697 U/ml Patient 140 34 28 6 72 85 7210 U/ml Patient 140 29 19 11 81 73 8112 U/ml Patient 140 29 15 11 85 73 8515 U/ml Patient 140 28 13 12 87 70 87Patient 140 14 10 26 90 35 90NMP22 (Matritech)Laboratory analysisSozen 1999 163No. of patients 140, ofwhom no previous historyof BC NS, history of BCNS, control group of 100with benign urologicaldisease or renal or prostatecancerCytology (VU orcatheterised)293© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Appendix 15294Specificity(%)Sensitivity(%)Numberanalysed TP FP FN TNStudy Test Cut-off Unit of analysisNMP22 (Matritech) > 6.4 U/ml Specimen 274 45 42 21 166 68 80Laboratory analysispTa 44 27 17 61pT1 6 6 0 100≥ pT2 6 5 1 83CIS 10 7 3 70G1 13 4 9 31G2 27 20 7 74G3 26 21 5 81Low risk (pTaG1, pTaG2) 39 23 16 59High risk (pTaG3, T1) 11 10 1 90Invasive (pT2–T4) 6 5 1 835 U/ml Specimen 274 48 59 18 149 73 727 U/ml Specimen 274 43 34 23 174 65 8410 U/ml Specimen 274 32 17 34 191 49 92Cytology (VU) Specimen 200 18 12 24 146 43 92Stampfer 1998 149No. of patients 231, ofwhom no previous historyof BC 0, history of BC 23112 U/ml Patient 669 28 124 20 497 58 80CIS–pT1 39 19 20 49pT2–T4 9 9 0 100G1 4 2 2 50G2 35 19 16 54G3 9 7 2 78< 10 mm 19 6 13 3210–30 mm 17 11 6 65> 30 mm 12 11 1 92Single 27 12 15 44Two to four 10 6 4 60Five or more 11 10 1 91NMP22 (Konica-Matritech)Laboratory analysisTakeuchi 2004 164No. of patients 669, ofwhom no previous historyof BC 48, history of BC 0,benign disease 621

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Specificity(%)Sensitivity(%)Numberanalysed TP FP FN TNStudy Test Cut-off Unit of analysisCytology (VU) Patient 669 21 0 27 621 44 100CIS–pT1 39 15 24 39pT2–T4 9 6 3 67G1 4 1 3 25G2 35 13 22 37G3 9 7 2 78< 10 mm 19 4 15 2110–30 mm 17 8 9 47> 30 mm 12 9 3 75Single 27 9 18 33Two to four 10 3 7 30Five or more 11 9 2 82Patient 48 29 19 60CIS–pT1 39 20 19 51pT2–T4 9 9 0 100G1 4 2 2 50G2 35 20 15 57G3 9 7 2 78< 10 mm 19 7 12 3710–30 mm 17 11 6 65> 30 mm 12 11 1 92Single 27 13 14 48Two to four 10 6 4 60Five or more 11 10 1 91NMP22 + cytology(VU)295© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Appendix 15296Specificity(%)Sensitivity(%)Numberanalysed TP FP FN TNStudy Test Cut-off Unit of analysis≥ 10 U/ml Patient 196 35 28 17 116 67 81pTa 24 14 10 58pT1 14 10 4 71pT2 8 6 2 75pT3 3 3 0 100CIS 1 1 0 100With well-differentiated 21 11 10 52tumoursWith poorlydifferentiated7 6 1 86tumoursWith moderatelydifferentiated22 17 5 77tumoursNo previous history of 69 9 7 1 52 90 88BCHistory of BC 127 27 21 15 64 64 75Non-invasive 39 25 14 64Muscle invasive 1 9 2 82Cystitis, inflammation, 58 17 41 71UTIBenign prostatic38 5 33 87hyperplasiaCalculi 44 7 37 84Non-TCC malignancies 2 2 0 0No urinary tract disease 51 2 49 96NMP22 (Matritech)BladderChekTalwar 2007 150No. of patients 196, ofwhom no previous historyof BC 69, history of BC 127

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Specificity(%)Sensitivity(%)Numberanalysed TP FP FN TNStudy Test Cut-off Unit of analysisCytology (VU) Patient 196 11 2 41 142 21 99pTa 24 14 10 58pT1 14 2 12 14pT2 8 2 6 25pT3 3 3 0 100CIS 1 0 1 0With well-differentiated 21 2 19 10tumoursWith poorlydifferentiated7 4 3 57tumoursWith moderatelydifferentiated22 4 18 18tumoursNo previous history of 69 1 0 9 59 10 100BCHistory of BC 127 9 0 33 85 21 100Non-invasive 39 5 34 13Muscle invasive 11 5 6 45Cystitis, inflammation, 58 1 57 98UTIBenign prostatic38 0 38 100hyperplasiaCalculi 44 0 44 100Non-TCC malignancies 2 1 1 50No urinary tract disease 51 0 51 100297© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Appendix 15298Specificity(%)Sensitivity(%)Numberanalysed TP FP FN TNStudy Test Cut-off Unit of analysisPatient 870 100 281 36 453 74 62pTa 65 51 14 79pT1 6 5 1 83≥ pT2 19 13 6 68CIS 14 13 1 93Patient 870 39 17 97 814 29 98pTa 65 8 57 12pT1 6 4 2 67≥ pT2 19 9 10 47CIS 14 7 7 50Patient 870 114 284 22 450 84 61pTa 65 51 14 79pT1 6 5 1 83≥ pT2 19 15 4 79CIS 14 14 0 100ImmunoCyt Presence of onegreen or one redfluorescent cellTetu 2005 119No. of patients 904, ofwhom no previous historyof BC NS, history of BC NSCytology (VU) (The studyincluded caseswith atypiassuspicious formalignancy in thenegative category)ImmunoCyt +cytology (VU)NMP22 (Matritech) 10 U/ml Patient 100 26 36 14 24 65 40BladderChekTumourG1–2 22 12 10 55G3 18 14 4 78pTa 22 12 10 55CIS 9 7 2 78Invasive T1 4 2 2 50Invasive ≥ T2 3 3 0 100Cytology (VU) Patient 85 15 11 19 40 44 78Cytology (BW) Patient 94 26 23 8 37 76 62TumourG1–2 22 17 5 83G3 18 15 3 83pTa 22 18 4 82CIS 9 7 2 78T1 4 3 1 75≥ T2 3 2 1 67Tritschler 2007 80No. of patients 100, ofwhom no previous historyof BC 30, history of BC 70

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Specificity(%)Sensitivity(%)Numberanalysed TP FP FN TNStudy Test Cut-off Unit of analysisNMP22 (Matritech) 10 U/ml Patient 291 44 62 47 138 48 69Laboratory analysispTa 47 23 24 49pT1 25 11 14 44≥ pT2 19 10 9 53G1 23 12 11 52G2 38 17 21 45G3 30 15 15 50Cytology (VU) Patient 291 54 0 37 200 59 100pTa 47 23 24 49pT1 25 16 9 64≥ pT2 19 15 4 79G1 23 4 19 17G2 38 23 15 61G3 30 27 3 90Cytology (BW) Patient 200 48 0 35 117 58 100pTa 47 22 25 47pT1 25 18 7 25≥ pT2 19 15 4 79G1 23 5 18 22G2 38 21 17 55G3 30 27 3 90Wiener 1998 151No. of patients 291, ofwhom no previous historyof BC 190, history of BC101299© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Appendix 15300Specificity(%)Sensitivity(%)Numberanalysed TP FP FN TNStudy Test Cut-off Unit of analysisPatient 250 25 56 14 155 64 73FISH (UroVysion) More than twochromosomalgains ofchromosomes 3,7 or 17 in at leastfour analysed cells,or homozygous9p21 deletionin at least 12analysed cells, orisolated trisomyof chromosome 3,7, or 17 in at least10% of analysedcellsYoder 2007 106No. of patients 250, ofwhom no previous historyof BC 0, history of BC 250> 10 U/ml Patient 330 18 45 0 267 100 86NMP22 (Matritech)Laboratory analysisZippe 1999 152Cytology (VU) Patient 330 6 0 12 312 33 100No. of patients 330, ofwhom no previous historyof BC 330, history of BC 0NMP22 > 10 U/ml Patient 146 8 14 0 124 100 90Cytology (VU) Patient 146 2 0 6 138 25 100[Zippe 1999 153 ]No. of patients 146, ofwhom no previous historyof BC 146, history of BC 0BC, bladder cancer; BW, bladder wash; FN, false negative; FP, false positive; NS, not stated; R, rigid; TN, true negative; TP, true positive; UTI, urinary tract infections; UUT, upper urinarytract; VU, voided urine.

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Appendix 16Cut-offs for a positive test usedin studies reporting FISHStudyDaniely 2007 94Friedrich 2003 95Halling 2000 97Junker 2006 98Kipp 2008 99May 2007 107Meiers 2007 100Mian 2003 101Moonen 2007 102Sarosdy 2002 108Sarosdy 2006 103Skacel 2003 104Sokolova 2000 105Yoder 2007 106Cut-offMinimum of four cells with gains of two or more chromosomes, or 12 or more cells with homozygousloss of the 9p21 locusIf 20% of the cells had a gain of two or more chromosomes (3, 7 or 17), or 40% of the cells had a gainof one chromosome or 40% loss of 9p21 locusFive or more cells with polysomyFive or more cells showed gains of more than one chromosome (3, 7 or 17), or 10 or more cellsshowed gains of a single chromosome (3, 7 or 17), or 10 or more cells showed homozygous loss of the9p21 locusFour or more cells had polysomic signal patterns (gain of two or more of the four chromosomes in anindividual cell), 10 or more cells demonstrated tetrasomy (four signal patterns for all four probes), or> 20% of the cells demonstrated 9p21 homozygous deletion (loss of the two 9p21 signals)Gain of two or more chromosomes in five or more cells per slide, or in cases of isolated gains ofchromosome 3, 7, or 17 when the proportion of cells with such a gain was 10% or more of at least100 cells evaluated, or when there were 10 or more cells with 9p21 lossChromosomal gain of two or more chromosomes (+3, +7, +17) in four or more cells, or deletion of9p21 in 12 or more cellsFour or more aneusomic of 25 counted cellsFour or more of the 25 morphologically abnormal cells showed gains of two or more chromosomes(3, 7 or 17), or 12 or more of the 25 cells had no 9p21 signalsAneuploidy of chromosomes 3, 7 and 17 or loss of the 9p21 locusAssay was performed according to product instructions [the UroVysion Bladder Cancer Kit(UroVysion Kit) is designed to detect aneuploidy for chromosomes 3, 7, 17, and loss of the 9p21 locus]Chromosomal gain of two or more chromosomes in five or more cells per slide, or in cases of isolatedgain of chromosome 3, 7 or 17 when the number of cells with such gain was ≥ 10%, or when 9p21 losswas the only abnormality, 12 or more cells with such lossFive or more cells with polysomyMore than two chromosomal gains of chromosomes 3, 7 or 17 in at least four analysed cells, orhomozygous 9p21 deletion in at least 12 analysed cells, or isolated trisomy of chromosome 3, 7, or 17in at least 10% of analysed cells301© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Appendix 17Model structureNon-muscleinvasiveNo tumour (one of risk groups)False-positive (one of risk groups)Recurrence (one of risk groups)Progression to muscle invasiveFalse-negative (one of risk groups)Low risk (no treat)Intermediate risk (no treat)High risk (no treat)Progression to muscle invasive (no treat)SurviveDieSurviveTrue-negativeFalse-positiveTrue-positiveFalse-negativeDieTrue-negativeFalse-positiveSurviveTrue-positiveFalse-negativeDieTrue-negativeFalse-positiveSurviveTrue-positiveFalse-negativeDieSurviveDieTrue-negativeFalse-positiveSurviveTrue-positiveFalse-negativeDieNot detectedDetectedNo progressProgressNo progressProgressNo progressProgressNo progressProgressNot detectedLow riskLow riskNo progress Intermediate riskDetectedHigh riskProgress to muscle invasiveNot detectedNo progressIntermediate riskIntermediate riskNo progressDetectedHigh riskProgress to muscle invasiveSurviveNot detectedHigh riskNo progressDetectedProgressNot detectedProgressSurviveDetectedDieDieNot detectedIntermediate riskIntermediate riskNo progressDetectedHigh riskNo progressProgressNot detectedHigh riskSurviveNo progressDetectedProgressNot detectedProgressSurviveDetectedDieDieNot detectedNo progressionNo progressDetectedSurviveProgressNot detectedProgressionDetectedDieSurviveDieSurviveDieSurviveDieNo tumour (one of risk groups)False-positive (one of risk groups)Recurrence (one of risk groups)Progression to muscle invasiveFalse-negative (one of risk groups)DeadNo tumour (one of risk groups)False-positive (one of risk groups)Recurrence (one of risk groups)Progression to muscle invasiveFalse-negative (one of risk groups)DeadNo tumour (one of risk groups)False-positive (one of risk groups)Recurrence (one of risk groups)Progression to muscle invasiveFalse-negative (one of risk groups)DeadProgression to muscle invasiveDeadNo tumour (one of risk groups)False-positive (one of risk groups)Recurrence (one of risk groups)Progression to muscle invasiveFalse-negative (one of risk groups)DeadLow risk (no treat)Recurrence (one of risk groups)Recurrence (one of risk groups)Recurrence (one of risk groups)Progression to muscle invasiveImmediate risk (no treat)Recurrence (one of risk groups)Recurrence (one of risk groups)Progression to muscle invasiveHigh risk (no treat)Recurrence (one of risk groups)Progression to muscle invasiveProgression to muscle invasive (no treat)Progression to muscle invasiveDeadDeadImmediate risk (no treat)Recurrence (one of risk groups)False-positive (one of risk groups)Progression to muscle invasiveHigh risk (no treat)Recurrence (one of risk groups)Progression to muscle invasiveProgression to muscle invasive (no treat)Progression to muscle invasiveDeadDeadHigh risk (no treat)Recurrence (one of risk groups)Progression to muscle invasiveProgression to muscle invasive (no treat)Progression to muscle invasiveDeadDeadProgression to muscle invasive (no treat)Progression to muscle invasiveDeadSurviveDeadDeadFIGURE 36 Diagram of Markov model for non-muscle-invasive disease.303© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Appendix 17Non-muscle invasiveLow riskIntermediate riskSurviveDieSurviveTrue-negativeFalse-positiveTrue-positiveFalse-negativeTrue-negativeFalse-positiveTrue-positiveFalse-negativeNo progressProgressNo progressProgressDieFirst testPositive thenPDD/WLCNegativePositiveNegative(+)SurviveDieSurviveDieTrue-positiveFalse-positiveFalse-negativeTrue-negativeMuscle invasiveSurviveDieSurviveDieSurviveDieHigh riskLocal muscle invasiveMetastasesNo progressProgress to muscleinvasiveSurviveDieSurviveDieSurviveDieNot detectedDetectedNot detectedDetectedTrue-negativeFalse-positiveTrue-positiveFalse-negativeNo tumourTumourLow riskIntermediate riskHigh riskLow riskIntermediate riskHigh riskNo progressProgressNo progressProgressFIGURE 37 Diagram of decision model.Muscle invasiveNo tumourRecurrenceProgress to metastasesDieSurviveDieSurviveDieSurviveDieNo tumour recurrenceTumour recurrenceNo tumour recurrenceTumour recurrenceNo progressProgressNo progressProgressNo tumourRecurrenceProgress to metastasesDeadNo tumourRecurrenceProgress to metastasesDeadProgress to metastasesDeadDead304FIGURE 38 Diagram of Markov model for muscle-invasive disease.

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Appendix 18Summary of studies reporting prognosisand all-cause mortality rates for the UK305© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Appendix 18306TABLE 55 Summary of studies reporting prognostic factors for recurrenceIntravesicalinstillationsTumourstatus No. of cases Grade T stage CIS Multiplicity Tumour sizeStudy Study type178 No Yes – No No –Loening 1980 186 Retrospective Primary andrecurrent468 No Yes Yes No Yes –Yes Yes – No No –Narayana Prospective Primary and1983 189 recurrent308 No – – Yes Yes –Dalesio 1983 182 Prospective Primary andrecurrentParmar 1989 191 Prospective Primary 305 No No – Yes No –Witjes 1992 194 Prospective Primary 1026 No Yes No No – YesKiemeney Prospective Primary 1674 No Yes No Yes – Yes1993 184Kiemeney Prospective Primary 1674 Yes Yes No – – Yes1994 185469 No No – Yes No –Witjes 1994 195 Prospective Primary andrecurrent371 No No – Yes No –Mulders 1994 188 Prospective Primary andrecurrent576 Yes No – Yes No –Kurth 1995 181 Retrospective Primary andrecurrent2535 – – – – – YesPawinski Meta-analysis Primary and1996 192 recurrentShinka 1997 193 Prospective Primary 141 No Yes No No Yes –Millán- Retrospective Primary 1529 No No Yes Yes Yes Yes2000 187RodriguezOosterlinck Guideline – – – – – – – –2001 1902596 Yes Yes Yes Yes Yes YesSylvester 2006 20 Retrospective Primary andrecurrent473 Yes No No Yes No YesGarcía 2006 183 Retrospective Primary andrecurrent

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4TABLE 56 Summary of studies reporting prognostic factors for progressionIntravesicalinstillationsTumourstatus No. of cases Grade T stage CIS Multiplicity Tumour sizeStudy Study type221 – Yes – – – –Herr 1997 196 – Primary andrecurrentKiemeney Prospective Primary 1674 Yes Yes Yes Yes – No1993 1841674 Yes Yes Yes – – NoKiemeney Prospective Primary and1994 185 recurrent576 Yes – – No Yes –Kurth 1995 181 Retrospective Primary andrecurrent2535 – – – – – NoPawinski Meta-analysis Primary and1996 192 recurrentMillán- Retrospective Primary 1529 Yes No Yes Yes Yes Yes2000 187Rodriguez2596 Yes Yes Yes Yes Yes YesSylvester 2006 20 Retrospective Primary andrecurrent473 Yes Yes Yes Yes No YesGarcía 2006 183 Retrospective Primary andrecurrent307© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Appendix 18TABLE 57 All-cause mortality rates for the UKAge (years) Female Male 30% female/70% male57 0.004643 0.007311 0.006510658 0.005050 0.007850 0.007010059 0.005639 0.008787 0.007842660 0.006160 0.010172 0.008968461 0.006807 0.011002 0.009743562 0.007443 0.012545 0.011014463 0.008116 0.013460 0.011856864 0.009152 0.015029 0.013265965 0.010041 0.016189 0.014344666 0.011114 0.017829 0.015814567 0.012173 0.019784 0.017500768 0.013430 0.021671 0.019198769 0.014893 0.024025 0.021285470 0.016138 0.026284 0.023240271 0.018145 0.029844 0.026334372 0.020737 0.032942 0.029280573 0.023061 0.036532 0.032490774 0.026217 0.041049 0.036599475 0.029660 0.045240 0.040566076 0.033232 0.050620 0.045403677 0.037046 0.056696 0.050801078 0.041599 0.062325 0.056107279 0.046364 0.069874 0.062821080 0.051959 0.076846 0.069379981 0.058465 0.085981 0.077726282 0.065710 0.094133 0.085606183 0.073339 0.103537 0.094477684 0.080283 0.111409 0.102071285 0.090944 0.121991 0.112676986 0.102260 0.136694 0.126363887 0.119838 0.159120 0.147335488 0.132897 0.174064 0.161713989 0.148659 0.192931 0.179649490 0.163740 0.201010 0.189829091 0.182212 0.220958 0.209334292 0.202965 0.243762 0.231522993 0.228008 0.269145 0.256803994 0.251579 0.281937 0.272829695 0.275949 0.319381 0.306351496 0.300473 0.342860 0.330143997 0.329979 0.371213 0.3588428308

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Appendix 19Results of cost–consequence analysis309© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Appendix 19TABLE 58 Ranking by diagnostic performanceRanking True negative True positive False positive False negative1 CTL_WLC (CTL_WLC) CSC_IMM_PDD (IMM_WLC)2 CTL_PDD (CTL_WLC) CSC_IMM_PDD (CSC_WLC)3 FISH_WLC (FISH_WLC) CSC_FISH_PDD (FISH_WLC)4 NMP22_WLC (NMP22_WLC)CSC_FISH_PDD (CSC_WLC)5 FISH_PDD (FISH_WLC) CSC_NMP22_PDD(NMP22_WLC)6 IMM_WLC (IMM_WLC) CSC_NMP22_PDD(CSC_WLC)CTL_WLC (CTL_WLC)CTL_PDD (CTL_WLC)FISH_WLC (FISH_WLC)NMP22_WLC (NMP22_WLC)FISH_PDD (FISH_WLC)IMM_WLC (IMM_WLC)CSC_IMM_PDD (IMM_WLC)CSC_IMM_PDD (CSC_WLC)CSC_FISH_PDD (FISH_WLC)CSC_FISH_PDD (CSC_WLC)CSC_NMP22_PDD(NMP22_WLC)CSC_NMP22_PDD(CSC_WLC)7 CSC_WLC (CSC_WLC) IMM_PDD (IMM_WLC) CSC_WLC (CSC_WLC) IMM_PDD (IMM_WLC)8 CSC_CTL_WLC (CSC_WLC)9 CSC_CTL_WLC (CTL_WLC)10 NMP22_PDD (NMP22_WLC)11 CSC_FISH_WLC (FISH_WLC)12 CSC_FISH_WLC (CSC_WLC)CSC_CTL_PDD (CSC_WLC)CSC_CTL_PDD (CTL_WLC)FISH_PDD (FISH_WLC)CSC_IMM_WLC (IMM_WLC)CSC_IMM_WLC (CSC_WLC)13 IMM_PDD (IMM_WLC) CSC_FISH_WLC (FISH_WLC)14 CSC_NMP22_WLC(NMP22_WLC)15 CSC_NMP22_WLC(CSC_WLC)CSC_FISH_WLC (CSC_WLC)CSC_PDD (CSC_WLC)16 CSC_PDD (CSC_WLC) CSC_NMP22_WLC(NMP22_WLC)17 CSC_IMM_WLC (IMM_WLC)18 CSC_IMM_WLC (CSC_WLC)19 CSC_CTL_PDD (CSC_WLC)20 CSC_CTL_PDD (CTL_WLC)21 CSC_FISH_PDD (FISH_WLC)22 CSC_FISH_PDD (CSC_WLC)23 CSC_NMP22_PDD(NMP22_WLC)24 CSC_NMP22_PDD(CSC_WLC)25 CSC_IMM_PDD (IMM_WLC)26 CSC_IMM_PDD (CSC_WLC)CSC_NMP22_WLC(CSC_WLC)NMP22_PDD (NMP22_WLC)IMM_WLC(IMM_WLC)CSC_CTL_WLC (CSC_WLC)CSC_CTL_WLC (CTL_WLC)FISH_WLC (FISH_WLC)CSC_WLC (CSC_WLC)NMP22_WLC (NMP22_WLC)CTL_PDD (CTL_WLC)CTL_WLC (CTL_WLC)CSC_CTL_WLC (CSC_WLC)CSC_CTL_WLC (CTL_WLC)NMP22_PDD (NMP22_WLC)CSC_FISH_WLC (FISH_WLC)CSC_FISH_WLC (CSC_WLC)IMM_PDD (IMM_WLC)CSC_NMP22_WLC(NMP22_WLC)CSC_NMP22_WLC(CSC_WLC)CSC_PDD (CSC_WLC)CSC_IMM_WLC (IMM_WLC)CSC_IMM_WLC (CSC_WLC)CSC_CTL_PDD (CSC_WLC)CSC_CTL_PDD (CTL_WLC)CSC_FISH_PDD (FISH_WLC)CSC_FISH_PDD (CSC_WLC)CSC_NMP22_PDD(NMP22_WLC)CSC_NMP22_PDD(CSC_WLC)CSC_IMM_PDD (IMM_WLC)CSC_IMM_PDD (CSC_WLC)CSC_CTL_PDD (CSC_WLC)CSC_CTL_PDD (CTL_WLC)FISH_PDD (FISH_WLC)CSC_IMM_WLC (IMM_WLC)CSC_IMM_WLC (CSC_WLC)CSC_FISH_WLC (FISH_WLC)CSC_FISH_WLC (CSC_WLC)CSC_PDD (CSC_WLC)CSC_NMP22_WLC(NMP22_WLC)CSC_NMP22_WLC(CSC_WLC)NMP22_PDD (NMP22_WLC)IMM_WLC (IMM_WLC)CSC_CTL_WLC (CSC_WLC)CSC_CTL_WLC (CTL_WLC)FISH_WLC (FISH_WLC)CSC_WLC (CSC_WLC)NMP22_WLC (NMP22_WLC)CTL_PDD (CTL_WLC)CTL_WLC (CTL_WLC)310For true results correct diagnosis and higher value life-years are better, and for false results incorrect diagnosis and lowervalue costs are better.

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4TABLE 59 Ranking by diagnostic performance and life-year and costRanking Correct diagnosis Incorrect diagnosis Life-years Cost1 CTL_WLC (CTL_WLC) CTL_WLC (CTL_WLC) CSC_IMM_PDD (CSC_WLC)2 CTL_PDD (CTL_WLC) CTL_PDD (CTL_WLC) CSC_IMM_PDD (IMM_WLC)3 FISH_WLC (FISH_WLC) FISH_WLC (FISH_WLC) CSC_FISH_PDD (CSC_WLC)4 FISH_PDD (FISH_WLC) FISH_PDD (FISH_WLC) CSC_FISH_PDD (FISH_WLC)5 NMP22_WLC (NMP22_WLC)NMP22_WLC (NMP22_WLC)CSC_NMP22_PDD(NMP22_WLC)6 IMM_WLC (IMM_WLC) IMM_WLC (IMM_WLC) CSC_NMP22_PDD(CSC_WLC)CTL_WLC (CTL_WLC)CTL_PDD (CTL_WLC)FISH_WLC (FISH_WLC)FISH_PDD (FISH_WLC)NMP22_WLC (NMP22_WLC)NMP22_PDD (NMP22_WLC)7 CSC_WLC (CSC_WLC) CSC_WLC (CSC_WLC) IMM_PDD (IMM_WLC) IMM_WLC (IMM_WLC)8 CSC_CTL_WLC (CSC_WLC)9 CSC_CTL_WLC (CTL_WLC)10 NMP22_PDD (NMP22_WLC)CSC_CTL_WLC (CSC_WLC)CSC_CTL_WLC (CTL_WLC)NMP22_PDD (NMP22_WLC)CSC_CTL_PDD (CTL_WLC)CSC_CTL_PDD (CSC_WLC)FISH_PDD (FISH_WLC)11 IMM_PDD (IMM_WLC) IMM_PDD (IMM_WLC) CSC_IMM_WLC (IMM_WLC)12 CSC_FISH_WLC (FISH_WLC)13 CSC_FISH_WLC (CSC_WLC)14 CSC_NMP22_WLC(NMP22_WLC)15 CSC_NMP22_WLC(CSC_WLC)CSC_FISH_WLC (FISH_WLC)CSC_FISH_WLC (CSC_WLC)CSC_NMP22_WLC(NMP22_WLC)CSC_NMP22_WLC(CSC_WLC)CSC_IMM_WLC (CSC_WLC)NMP22_PDD (NMP22_WLC)CSC_FISH_WLC (CSC_WLC)CSC_FISH_WLC (FISH_WLC)IMM_PDD (IMM_WLC)CSC_CTL_WLC (CTL_WLC)CSC_FISH_WLC (FISH_WLC)CSC_NMP22_WLC(NMP22_WLC)CSC_CTL_PDD (CTL_WLC)CSC_WLC (CSC_WLC)CSC_IMM_WLC (IMM_WLC)CSC_CTL_WLC (CSC_WLC)16 CSC_PDD (CSC_WLC) CSC_PDD (CSC_WLC) CSC_PDD (CSC_WLC) CSC_FISH_WLC (CSC_WLC)17 CSC_IMM_WLC (IMM_WLC)18 CSC_IMM_WLC (CSC_WLC)19 CSC_CTL_PDD (CSC_WLC)20 CSC_CTL_PDD (CTL_WLC)21 CSC_FISH_PDD (FISH_WLC)22 CSC_FISH_PDD (CSC_WLC)23 CSC_NMP22_PDD(NMP22_WLC)24 CSC_NMP22_PDD(CSC_WLC)25 CSC_IMM_PDD (IMM_WLC)26 CSC_IMM_PDD (CSC_WLC)CSC_IMM_WLC (IMM_WLC)CSC_IMM_WLC (CSC_WLC)CSC_CTL_PDD (CSC_WLC)CSC_CTL_PDD (CTL_WLC)CSC_FISH_PDD (FISH_WLC)CSC_FISH_PDD (CSC_WLC)CSC_NMP22_PDD(NMP22_WLC)CSC_NMP22_PDD(CSC_WLC)CSC_IMM_PDD (IMM_WLC)CSC_IMM_PDD (CSC_WLC)CSC_PDD (CSC_PDD)IMM_WLC (IMM_WLC)CSC_NMP22_WLC(NMP22_WLC)CSC_NMP22_WLC(CSC_WLC)CSC_CTL_WLC (CTL_WLC)CSC_CTL_WLC (CSC_WLC)FISH_WLC (FISH_WLC)NMP22_WLC (NMP22_WLC)CSC_WLC (CSC_WLC)CTL_PDD (CTL_WLC)CSC_FISH_PDD (FISH_WLC)CSC_NMP22_WLC(CSC_WLC)CSC_PDD (CSC_WLC)CSC_IMM_WLC (CSC_WLC)CSC_NMP22_PDD(NMP22_WLC)CSC_CTL_PDD (CSC_WLC)CSC_IMM_PDD (IMM_WLC)CSC_FISH_PDD (CSC_WLC)CSC_NMP22_PDD(CSC_WLC)CSC_IMM_PDD (CSC_WLC)311© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Appendix 20Cost-effectiveness acceptability curvesfor the eight strategies for changes inthe incidence rate (base case = 5%)Probability cost-effective1.00.90.80.70.60.50.40.30.2+ + + + + + + ++ + + +0.1+++ + + + + + + + ++ + ++ ++ + + + +0.0 + +++0 10 20 30 40 50++ + +++++ + + + ++++++++++Ceiling ratio (Rc) (£000)++++++++++CTL-WLC (CTL-WLC)CTL-PDD (CTL-WLC)FISH-PDD (FISH-WLC)IMM-PDD (IMM-WLC)CSC-FISH-WLC (FISH-WLC)CSC-PDD (CSC-WLC)CSC-IMM-PDD (IMM-WLC)CSC-IMM-PDD (CSC-WLC)FIGURE 39 Incidence rate is 1%.Probability cost-effective1.00.90.80.70.60.50.40.3+ + + + + + + + + + + + + + +0.20.1+ ++ + + ++ ++ + + + + + + + + ++ ++ + + +0.00 10 20 30 40 50+ + + + + + + +++ ++ ++++++++Ceiling ratio (Rc) (£000)++++++++++CTL-WLC (CTL-WLC)CTL-PDD (CTL-WLC)FISH-PDD (FISH-WLC)IMM-PDD (IMM-WLC)CSC-FISH-WLC (FISH-WLC)CSC-PDD (CSC-WLC)CSC-IMM-PDD (IMM-WLC)CSC-IMM-PDD (CSC-WLC)FIGURE 40 Incidence rate is 10%.313© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Appendix 20Probability cost-effective1.00.90.80.70.60.50.40.30.2++ + + + + + + + + + + + + + +++0.1+ + + + + + + ++ + + + +++ + + + + ++ +0.00 10 20 30 40 50++ + + + + + + + +++++++++++Ceiling ratio (Rc) (£000)++++++++++CTL-WLC (CTL-WLC)CTL-PDD (CTL-WLC)FISH-PDD (FISH-WLC)IMM-PDD (IMM-WLC)CSC-FISH-WLC (FISH-WLC)CSC-PDD (CSC-WLC)CSC-IMM-PDD (IMM-WLC)CSC-IMM-PDD (CSC-WLC)FIGURE 41 Incidence rate is 20%.314

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Appendix 21Cost-effectiveness acceptability curves forchanges to the performance of flexiblecystoscopy (base-case flexible cystoscopy isthe same as white light rigid cystoscopy)Probability cost-effective1.00.90.80.70.60.50.40.30.2+ + + + + + ++ + + + + ++ +0.1+ + + + + + + + ++ + + ++ + + +0.0 + + + ++ + 0 10 20 30 40 50+++++ ++++++++++++++++Ceiling ratio (Rc) (£000)++++++++++CTL-WLC (CTL-WLC)CTL-PDD (CTL-WLC)FISH-PDD (FISH-WLC)IMM-PDD (IMM-WLC)CSC-FISH-WLC (FISH-WLC)CSC-PDD (CSC-WLC)CSC-IMM-PDD (IMM-WLC)CSC-IMM-PDD (CSC-WLC)FIGURE 42 Sensitivity and specificity of flexible cystoscopy are increased by 5% from base case.Probability cost-effective1.00.90.80.70.60.50.40.30.2+ + + + + + + ++ + + + + +0.1++ + + + + + + ++ + + + +++ + +0.0 + + + ++ + 0 10 20 30 40 50+++++++ + + + +++++++++++Ceiling ratio (Rc) (£000)++++++++++CTL-WLC (CTL-WLC)CTL-PDD (CTL-WLC)FISH-PDD (FISH-WLC)IMM-PDD (IMM-WLC)CSC-FISH-WLC (FISH-WLC)CSC-PDD (CSC-WLC)CSC-IMM-PDD (IMM-WLC)CSC-IMM-PDD (CSC-WLC)FIGURE 43 Sensitivity and specificity of flexible cystoscopy are increased by 10% from base case.315© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Appendix 21Probability cost-effective1.00.90.80.70.60.50.40.30.2+ + + + + + + +++ ++ ++ ++ +0.1+ +++0.00 10 20 30 40 50+ + +++ + ++ + + ++++ ++++++++++Ceiling ratio (Rc) (£000)++++++++++CTL-WLC (CTL-WLC)CTL-PDD (CTL-WLC)FISH-PDD (FISH-WLC)IMM-PDD (IMM-WLC)CSC-FISH-WLC (FISH-WLC)CSC-PDD (CSC-WLC)CSC-IMM-PDD (IMM-WLC)CSC-IMM-PDD (CSC-WLC)FIGURE 44 Sensitivity and specificity of flexible cystoscopy are increased by 25% from base case.316

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Appendix 22Cost-effectiveness acceptability curves forchanges to the relative risk (RR) of progressionof bladder cancer for no treatment ofbladder cancer compared with treatmentof bladder cancer (base-case RR = 2.56)Probability cost-effective1.00.90.80.70.60.50.40.30.2+ + + + + + + + + + ++ +0.1+ + + + + + + + + + ++ + + + ++ + ++ +0.0 + + + + 0 10 20 30 40 50+++++ + + + +++++++++++++Ceiling ratio (Rc) (£000)++++++++++CTL-WLC (CTL-WLC)CTL-PDD (CTL-WLC)FISH-PDD (FISH-WLC)IMM-PDD (IMM-WLC)CSC-FISH-WLC (FISH-WLC)CSC-PDD (CSC-WLC)CSC-IMM-PDD (IMM-WLC)CSC-IMM-PDD (CSC-WLC)FIGURE 45 The relative risk for progression comparing no treatment with treatment is decreased to 2.0.Probability cost-effective1.00.90.80.70.60.50.40.30.2+ + + + + ++ + + + + + + + +0.1+ + + + + + + + ++ + + ++ ++ + + +0.0 + + + + 0 10 20 30 40 50+++++++++++++++++++++Ceiling ratio (Rc) (£000)++++++++++CTL-WLC (CTL-WLC)CTL-PDD (CTL-WLC)FISH-PDD (FISH-WLC)IMM-PDD (IMM-WLC)CSC-FISH-WLC (FISH-WLC)CSC-PDD (CSC-WLC)CSC-IMM-PDD (IMM-WLC)CSC-IMM-PDD (CSC-WLC)FIGURE 46 The relative risk for progression comparing no treatment with treatment is decreased to 1.5.317© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Appendix 22Probability cost-effective1.00.90.80.70.60.50.40.30.2+ + + + + ++ + + + + ++ ++0.1+ + + + + + + ++ + + + + + ++ + + +0.0 + + + + 0 10 20 30 40 50+++++ + + ++++++++++++++Ceiling ratio (Rc) (£000)++++++++++CTL-WLC (CTL-WLC)CTL-PDD (CTL-WLC)FISH-PDD (FISH-WLC)IMM-PDD (IMM-WLC)CSC-FISH-WLC (FISH-WLC)CSC-PDD (CSC-WLC)CSC-IMM-PDD (IMM-WLC)CSC-IMM-PDD (CSC-WLC)FIGURE 47 The relative risk for progression comparing no treatment with treatment is decreased to 1.0.318

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Appendix 23Cost-effectiveness acceptability curvesfor the eight strategies for changes in therelative risk (RR) for recurrence comparingPDD with WLC (base-case RR = 1)Probability cost-effective1.00.90.80.70.60.50.40.30.2+ + + + + + + ++ + + + ++ +0.1+ + + + + + + + ++ + + +++ + +0.0 + + + ++ + 0 10 20 30 40 50++++++ +++++++++++++++Ceiling ratio (Rc) (£000)++++++++++CTL-WLC (CTL-WLC)CTL-PDD (CTL-WLC)FISH-PDD (FISH-WLC)IMM-PDD (IMM-WLC)CSC-FISH-WLC (FISH-WLC)CSC-PDD (CSC-WLC)CSC-IMM-PDD (IMM-WLC)CSC-IMM-PDD (CSC-WLC)FIGURE 48 The relative risk for recurrence for the comparison of PDD with WLC is 0.9.Probability cost-effective1.00.90.80.70.60.50.40.30.2+ + + + + + + + +0.1+ + + + + + + + ++ + ++ + ++ + + + +0.0 + + + ++ + 0 10 20 30 40 50+++ + + + + ++++ ++ ++++++++Ceiling ratio (Rc) (£000)++++++++++CTL-WLC (CTL-WLC)CTL-PDD (CTL-WLC)FISH-PDD (FISH-WLC)IMM-PDD (IMM-WLC)CSC-FISH-WLC (FISH-WLC)CSC-PDD (CSC-WLC)CSC-IMM-PDD (IMM-WLC)CSC-IMM-PDD (CSC-WLC)FIGURE 49 The relative risk for recurrence for the comparison of PDD with WLC is 0.8.319© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Appendix 23Probability cost-effective1.00.90.80.70.60.50.40.30.2+ + + + + + + + + + + + + +0.1+ ++ + + + + + + ++ + + + + ++ ++0.0 + + + + +0 10 20 30 40 50++ + +++ + +++++++++++++++Ceiling ratio (Rc) (£000)++++++++++CTL-WLC (CTL-WLC)CTL-PDD (CTL-WLC)FISH-PDD (FISH-WLC)IMM-PDD (IMM-WLC)CSC-FISH-WLC (FISH-WLC)CSC-PDD (CSC-WLC)CSC-IMM-PDD (IMM-WLC)CSC-IMM-PDD (CSC-WLC)FIGURE 50 The relative risk for recurrence for the comparison of PDD with WLC is 0.64.320

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Appendix 24Cost-effectiveness acceptability curvesfor the eight strategies for changes in therelative risk (RR) for progression comparingPDD with WLC (base-case RR = 1)Probability cost-effective1.00.90.80.70.60.50.40.30.2+ + + + + ++ + + + + + + ++0.1+ + + + + + + + ++ + ++ +++ + +0.0 + + + + + 0 10 20 30 40 50++++ + + + ++++++++++++++Ceiling ratio (Rc) (£000)++++++++++CTL-WLC (CTL-WLC)CTL-PDD (CTL-WLC)FISH-PDD (FISH-WLC)IMM-PDD (IMM-WLC)CSC-FISH-WLC (FISH-WLC)CSC-PDD (CSC-WLC)CSC-IMM-PDD (IMM-WLC)CSC-IMM-PDD (CSC-WLC)FIGURE 51 The relative risk for progression for the comparison of PDD with WLC is 0.9.Probability cost-effective1.00.90.80.70.60.50.40.30.2+ + + + + + + + + + + + ++ +0.1+ + + + + + + + ++ + + + ++ + + +0.0 + + + + + 0 10 20 30 40 50+++ + + + + ++++ ++ ++++++++Ceiling ratio (Rc) (£000)++++++++++CTL-WLC (CTL-WLC)CTL-PDD (CTL-WLC)FISH-PDD (FISH-WLC)IMM-PDD (IMM-WLC)CSC-FISH-WLC (FISH-WLC)CSC-PDD (CSC-WLC)CSC-IMM-PDD (IMM-WLC)CSC-IMM-PDD (CSC-WLC)FIGURE 52 The relative risk for progression for the comparison of PDD with WLC is 0.8.321© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Appendix 24Probability cost-effective1.00.90.80.70.60.50.40.30.2+ + + + + ++ + + + + ++ +0.1+ + + + + + + + + ++ + + ++ ++0.0 + + + + + + + 0 10 20 30 40 50+++++ ++++++++++++++++Ceiling ratio (Rc) (£000)++++++++++CTL-WLC (CTL-WLC)CTL-PDD (CTL-WLC)FISH-PDD (FISH-WLC)IMM-PDD (IMM-WLC)CSC-FISH-WLC (FISH-WLC)CSC-PDD (CSC-WLC)CSC-IMM-PDD (IMM-WLC)CSC-IMM-PDD (CSC-WLC)FIGURE 53 The relative risk for progression for the comparison of PDD with WLC is 0.56.322

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Appendix 25Cost-effectiveness acceptability curves for theeight strategies for changes in the discountrate (base-case discount rate = 3.5%)Probability cost-effective1.00.90.80.70.60.50.40.3+ + + + + +0.2+ ++ + + + + + +0.1+ + + + + + + + ++ + + + + + +++ + +0.0 + + + 0 10 20 30 40 50+++ + + + +++++ ++++++++++Ceiling ratio (Rc) (£000)++++++++++CTL-WLC (CTL-WLC)CTL-PDD (CTL-WLC)FISH-PDD (FISH-WLC)IMM-PDD (IMM-WLC)CSC-FISH-WLC (FISH-WLC)CSC-PDD (CSC-WLC)CSC-IMM-PDD (IMM-WLC)CSC-IMM-PDD (CSC-WLC)FIGURE 54 The discount rate is 6%.Probability cost-effective1.00.90.80.70.60.50.40.30.2+ + + + + + + + + + + + ++0.1++ + + + + + + + + ++ + + ++ + + +++0.0 + + + 0 10 20 30 40 50Ceiling ratio (Rc) (£000)+++ + + + +++++ ++ ++ ++++++++++++++++CTL-WLC (CTL-WLC)CTL-PDD (CTL-WLC)FISH-PDD (FISH-WLC)IMM-PDD (IMM-WLC)CSC-FISH-WLC (FISH-WLC)CSC-PDD (CSC-WLC)CSC-IMM-PDD (IMM-WLC)CSC-IMM-PDD (CSC-WLC)FIGURE 55 The discount rate is 1%.323© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Appendix 25Probability cost-effective1.00.90.80.70.60.50.40.30.2+ + + + + + + + + + + + + + ++0.1+ + + + + + + + ++ + + ++ + + ++ + + + +0.0 + + 0 10 20 30 40 50+ +++ + ++ +++++++++++++++Ceiling ratio (Rc) (£000)++++++++++CTL-WLC (CTL-WLC)CTL-PDD (CTL-WLC)FISH-PDD (FISH-WLC)IMM-PDD (IMM-WLC)CSC-FISH-WLC (FISH-WLC)CSC-PDD (CSC-WLC)CSC-IMM-PDD (IMM-WLC)CSC-IMM-PDD (CSC-WLC)FIGURE 56 The discount rate is 0%.324

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Appendix 26Cost-effectiveness acceptability curves for theeight strategies for changes in proportionsin the risk groups for non-invasive disease(base case: proportion in low-risk group is 0.1and proportion is high-risk group is 0.45)Probability cost-effective1.00.90.80.70.60.50.40.30.2+ + + + + + + + + + + ++ + +0.1+ + + + + + + ++ ++ + + + + ++0.0 + + + ++ + 0 10 20 30 40 50+++++ ++++++++++++++++Ceiling ratio (Rc) (£000)++++++++++CTL-WLC (CTL-WLC)CTL-PDD (CTL-WLC)FISH-PDD (FISH-WLC)IMM-PDD (IMM-WLC)CSC-FISH-WLC (FISH-WLC)CSC-PDD (CSC-WLC)CSC-IMM-PDD (IMM-WLC)CSC-IMM-PDD (CSC-WLC)FIGURE 57 Proportions in the high- and low-risk groups are 30%.Probability cost-effective1.00.90.80.70.60.50.40.30.2+ + + + + + + + + + + + + +0.1+ + ++ + + + + + + ++ + + + + ++ + ++ + + ++0.00 10 20 30 40 50+++ + + + +++++++++++++++Ceiling ratio (Rc) (£000)++++++++++CTL-WLC (CTL-WLC)CTL-PDD (CTL-WLC)FISH-PDD (FISH-WLC)IMM-PDD (IMM-WLC)CSC-FISH-WLC (FISH-WLC)CSC-PDD (CSC-WLC)CSC-IMM-PDD (IMM-WLC)CSC-IMM-PDD (CSC-WLC)FIGURE 58 Proportions in the high- and low-risk groups are 10% and 60% respectively.325© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Appendix 27Cost-effectiveness acceptability curvesfor the eight strategies for changes inthe starting age and time horizonProbability cost-effective1.00.90.80.70.60.50.40.3+ + + + + + +0.2++ + + + +++0.1+ + + + + + + + ++ + + ++ + + +0.0 + + + ++ + 0 10 20 30 40 50+++ + + + ++ +++++++++++++Ceiling ratio (Rc) (£000)++++++++++CTL-WLC (CTL-WLC)CTL-PDD (CTL-WLC)FISH-PDD (FISH-WLC)IMM-PDD (IMM-WLC)CSC-FISH-WLC (FISH-WLC)CSC-PDD (CSC-WLC)CSC-IMM-PDD (IMM-WLC)CSC-IMM-PDD (CSC-WLC)FIGURE 59 Starting age is 57 years.Probability cost-effective1.00.90.80.70.60.50.40.30.2+ + + + + + + + + + + + + + +0.1++ + + + + + + ++ + ++ + + + +0.0 + + + ++ + 0 10 20 30 40 50++++ + + +++++++++++++++Ceiling ratio (Rc) (£000)++++++++++CTL-WLC (CTL-WLC)CTL-PDD (CTL-WLC)FISH-PDD (FISH-WLC)IMM-PDD (IMM-WLC)CSC-FISH-WLC (FISH-WLC)CSC-PDD (CSC-WLC)CSC-IMM-PDD (IMM-WLC)CSC-IMM-PDD (CSC-WLC)FIGURE 60 Starting age is 77 years.327© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Appendix 27Probability cost-effective1.00.90.80.70.60.50.40.30.2+ + + + + +++0.1+ + + + + + + ++ + + + + + +0.0 + + + + + + + + + + ++ + +0 10 20 30 40 50++++ + + + +++ ++++++++++++Ceiling ratio (Rc) (£000)++++++++++CTL-WLC (CTL-WLC)CTL-PDD (CTL-WLC)FISH-PDD (FISH-WLC)IMM-PDD (IMM-WLC)CSC-FISH-WLC (FISH-WLC)CSC-PDD (CSC-WLC)CSC-IMM-PDD (IMM-WLC)CSC-IMM-PDD (CSC-WLC)FIGURE 61 Time horizon is 10 years.328

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Appendix 28Cost-effectiveness acceptability curves forthe eight strategies when WLC is replacedby PDD in follow-up for each strategyProbability cost-effective1.00.90.80.70.60.50.40.3+ + + + + + + + + + + + + + + +0.20.1+ + + + + + + + ++ + ++ + + ++ ++0.0 + + + + 0 10 20 30 40 50+++ + + + +++++++++++++++Ceiling ratio (Rc) (£000)++++++++++CTL-WLC (CTL-WLC)CTL-PDD (CTL-WLC)FISH-PDD (FISH-WLC)IMM-PDD (IMM-WLC)CSC-FISH-WLC (FISH-WLC)CSC-PDD (CSC-WLC)CSC-IMM-PDD (IMM-WLC)CSC-IMM-PDD (CSC-WLC)329© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Appendix 29Cost-effectiveness acceptability curvesfor the eight strategies when qualityof life measures are incorporated toproduce quality-adjusted life-yearsProbability cost-effective1.00.90.80.70.60.50.40.30.2++ + + + + + + + + + + + + + +0.1+ + + + + + + + + ++ + + + + ++0.0 + + + ++ + 0 10 20 30 40 50+++ + + + +++++ ++++++++++Ceiling ratio (Rc) (£000)++++++++++CTL-WLC (CTL-WLC)CTL-PDD (CTL-WLC)FISH-PDD (FISH-WLC)IMM-PDD (IMM-WLC)CSC-FISH-WLC (FISH-WLC)CSC-PDD (CSC-WLC)CSC-IMM-PDD (IMM-WLC)CSC-IMM-PDD (CSC-WLC)331© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Volume 1, 1997Health Technology Assessment reportspublished to dateNo. 1Home parenteral nutrition: a systematicreview.By Richards DM, Deeks JJ, SheldonTA, Shaffer JL.No. 2Diagnosis, management and screeningof early localised prostate cancer.A review by Selley S, Donovan J,Faulkner A, Coast J, Gillatt D.No. 3The diagnosis, management, treatmentand costs of prostate cancer in Englandand Wales.A review by Chamberlain J, Melia J,Moss S, Brown J.No. 4Screening for fragile X syndrome.A review by Murray J, Cuckle H,Taylor G, Hewison J.No. 5A review of near patient testing inprimary care.By Hobbs FDR, Delaney BC,Fitzmaurice DA, Wilson S, Hyde CJ,Thorpe GH, et al.No. 6Systematic review of outpatient servicesfor chronic pain control.By McQuay HJ, Moore RA, EcclestonC, Morley S, de C Williams AC.No. 7Neonatal screening for inborn errors ofmetabolism: cost, yield and outcome.A review by Pollitt RJ, Green A,McCabe CJ, Booth A, Cooper NJ,Leonard JV, et al.No. 8Preschool vision screening.A review by Snowdon SK,Stewart-Brown SL.No. 9Implications of socio-cultural contextsfor the ethics of clinical trials.A review by Ashcroft RE, ChadwickDW, Clark SRL, Edwards RHT, Frith L,Hutton JL.No. 10A critical review of the role of neonatalhearing screening in the detection ofcongenital hearing impairment.By Davis A, Bamford J, Wilson I,Ramkalawan T, Forshaw M, Wright S.No. 11Newborn screening for inborn errors ofmetabolism: a systematic review.By Seymour CA, Thomason MJ,Chalmers RA, Addison GM, Bain MD,Cockburn F, et al.No. 12Routine preoperative testing: asystematic review of the evidence.By Munro J, Booth A, Nicholl J.No. 13Systematic review of the effectiveness oflaxatives in the elderly.By Petticrew M, Watt I, Sheldon T.No. 14When and how to assess fast-changingtechnologies: a comparative study ofmedical applications of four generictechnologies.A review by Mowatt G, Bower DJ,Brebner JA, Cairns JA, Grant AM,McKee L.Volume 2, 1998No. 1Antenatal screening for Down’ssyndrome.A review by Wald NJ, Kennard A,Hackshaw A, McGuire A.No. 2Screening for ovarian cancer: asystematic review.By Bell R, Petticrew M, Luengo S,Sheldon TA.No. 3Consensus development methods,and their use in clinical guidelinedevelopment.A review by Murphy MK, Black NA,Lamping DL, McKee CM, SandersonCFB, Askham J, et al.No. 4A cost–utility analysis of interferon betafor multiple sclerosis.By Parkin D, McNamee P, Jacoby A,Miller P, Thomas S, Bates D.No. 5Effectiveness and efficiency of methodsof dialysis therapy for end-stage renaldisease: systematic reviews.By MacLeod A, Grant A, DonaldsonC, Khan I, Campbell M, Daly C, et al.No. 6Effectiveness of hip prostheses inprimary total hip replacement: a criticalreview of evidence and an economicmodel.By Faulkner A, Kennedy LG, BaxterK, Donovan J, Wilkinson M, Bevan G.No. 7Antimicrobial prophylaxis in colorectalsurgery: a systematic review ofrandomised controlled trials.By Song F, Glenny AM.No. 8Bone marrow and peripheralblood stem cell transplantation formalignancy.A review by Johnson PWM,Simnett SJ, Sweetenham JW, Morgan GJ,Stewart LA.No. 9Screening for speech and languagedelay: a systematic review of theliterature.By Law J, Boyle J, Harris F,Harkness A, Nye C.No. 10Resource allocation for chronicstable angina: a systematicreview of effectiveness, costs andcost-effectiveness of alternativeinterventions.By Sculpher MJ, Petticrew M,Kelland JL, Elliott RA, Holdright DR,Buxton MJ.No. 11Detection, adherence and control ofhypertension for the prevention ofstroke: a systematic review.By Ebrahim S.No. 12Postoperative analgesia and vomiting,with special reference to day-casesurgery: a systematic review.By McQuay HJ, Moore RA.No. 13Choosing between randomised andnonrandomised studies: a systematicreview.By Britton A, McKee M, Black N,McPherson K, Sanderson C, Bain C.No. 14Evaluating patient-based outcomemeasures for use in clinical trials.A review by Fitzpatrick R, Davey C,Buxton MJ, Jones DR.333© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Health Technology Assessment reports published to date334No. 15Ethical issues in the design and conductof randomised controlled trials.A review by Edwards SJL, Lilford RJ,Braunholtz DA, Jackson JC, Hewison J,Thornton J.No. 16Qualitative research methods in healthtechnology assessment: a review of theliterature.By Murphy E, Dingwall R,Greatbatch D, Parker S, Watson P.No. 17The costs and benefits of paramedicskills in pre-hospital trauma care.By Nicholl J, Hughes S, Dixon S,Turner J, Yates D.No. 18Systematic review of endoscopicultrasound in gastro-oesophagealcancer.By Harris KM, Kelly S, Berry E,Hutton J, Roderick P, Cullingworth J,et al.No. 19Systematic reviews of trials and otherstudies.By Sutton AJ, Abrams KR, Jones DR,Sheldon TA, Song F.No. 20Primary total hip replacement surgery:a systematic review of outcomesand modelling of cost-effectivenessassociated with different prostheses.A review by Fitzpatrick R, ShortallE, Sculpher M, Murray D, Morris R,Lodge M, et al.Volume 3, 1999No. 1Informed decision making: anannotated bibliography and systematicreview.By Bekker H, Thornton JG,Airey CM, Connelly JB, Hewison J,Robinson MB, et al.No. 2Handling uncertainty when performingeconomic evaluation of healthcareinterventions.A review by Briggs AH, Gray AM.No. 3The role of expectancies in the placeboeffect and their use in the delivery ofhealth care: a systematic review.By Crow R, Gage H, Hampson S,Hart J, Kimber A, Thomas H.No. 4A randomised controlled trial ofdifferent approaches to universalantenatal HIV testing: uptake andacceptability. Annex: Antenatal HIVtesting – assessment of a routinevoluntary approach.By Simpson WM, Johnstone FD,Boyd FM, Goldberg DJ, Hart GJ,Gormley SM, et al.No. 5Methods for evaluating area-wide andorganisation-based interventions inhealth and health care: a systematicreview.By Ukoumunne OC, Gulliford MC,Chinn S, Sterne JAC, Burney PGJ.No. 6Assessing the costs of healthcaretechnologies in clinical trials.A review by Johnston K, Buxton MJ,Jones DR, Fitzpatrick R.No. 7Cooperatives and their primary careemergency centres: organisation andimpact.By Hallam L, Henthorne K.No. 8Screening for cystic fibrosis.A review by Murray J, Cuckle H,Taylor G, Littlewood J, Hewison J.No. 9A review of the use of health statusmeasures in economic evaluation.By Brazier J, Deverill M, Green C,Harper R, Booth A.No. 10Methods for the analysis of qualityof-lifeand survival data in healthtechnology assessment.A review by Billingham LJ,Abrams KR, Jones DR.No. 11Antenatal and neonatalhaemoglobinopathy screening in theUK: review and economic analysis.By Zeuner D, Ades AE, Karnon J,Brown J, Dezateux C, Anionwu EN.No. 12Assessing the quality of reports ofrandomised trials: implications for theconduct of meta-analyses.A review by Moher D, Cook DJ,Jadad AR, Tugwell P, Moher M,Jones A, et al.No. 13‘Early warning systems’ for identifyingnew healthcare technologies.By Robert G, Stevens A, Gabbay J.No. 14A systematic review of the role ofhuman papillomavirus testing within acervical screening programme.By Cuzick J, Sasieni P, Davies P,Adams J, Normand C, Frater A, et al.No. 15Near patient testing in diabetes clinics:appraising the costs and outcomes.By Grieve R, Beech R, Vincent J,Mazurkiewicz J.No. 16Positron emission tomography:establishing priorities for healthtechnology assessment.A review by Robert G, Milne R.No. 17 (Pt 1)The debridement of chronic wounds: asystematic review.By Bradley M, Cullum N, Sheldon T.No. 17 (Pt 2)Systematic reviews of wound caremanagement: (2) Dressings and topicalagents used in the healing of chronicwounds.By Bradley M, Cullum N, Nelson EA,Petticrew M, Sheldon T, Torgerson D.No. 18A systematic literature review ofspiral and electron beam computedtomography: with particular referenceto clinical applications in hepaticlesions, pulmonary embolus andcoronary artery disease.By Berry E, Kelly S, Hutton J,Harris KM, Roderick P, Boyce JC, et al.No. 19What role for statins? A review andeconomic model.By Ebrahim S, Davey SmithG, McCabe C, Payne N, Pickin M,Sheldon TA, et al.No. 20Factors that limit the quality, numberand progress of randomised controlledtrials.A review by Prescott RJ, Counsell CE,Gillespie WJ, Grant AM, Russell IT,Kiauka S, et al.No. 21Antimicrobial prophylaxis in total hipreplacement: a systematic review.By Glenny AM, Song F.No. 22Health promoting schools and healthpromotion in schools: two systematicreviews.By Lister-Sharp D, Chapman S,Stewart-Brown S, Sowden A.No. 23Economic evaluation of a primarycare-based education programme forpatients with osteoarthritis of the knee.A review by Lord J, Victor C,Littlejohns P, Ross FM, Axford JS.

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Volume 4, 2000No. 1The estimation of marginal timepreference in a UK-wide sample(TEMPUS) project.A review by Cairns JA,van der Pol MM.No. 2Geriatric rehabilitation followingfractures in older people: a systematicreview.By Cameron I, Crotty M, Currie C,Finnegan T, Gillespie L, Gillespie W,et al.No. 3Screening for sickle cell disease andthalassaemia: a systematic review withsupplementary research.By Davies SC, Cronin E, Gill M,Greengross P, Hickman M, Normand C.No. 4Community provision of hearing aidsand related audiology services.A review by Reeves DJ, Alborz A,Hickson FS, Bamford JM.No. 5False-negative results in screeningprogrammes: systematic review ofimpact and implications.By Petticrew MP, Sowden AJ,Lister-Sharp D, Wright K.No. 6Costs and benefits of communitypostnatal support workers: arandomised controlled trial.By Morrell CJ, Spiby H, Stewart P,Walters S, Morgan A.No. 7Implantable contraceptives (subdermalimplants and hormonally impregnatedintrauterine systems) versus otherforms of reversible contraceptives: twosystematic reviews to assess relativeeffectiveness, acceptability, tolerabilityand cost-effectiveness.By French RS, Cowan FM,Mansour DJA, Morris S, Procter T,Hughes D, et al.No. 8An introduction to statistical methodsfor health technology assessment.A review by White SJ, Ashby D,Brown PJ.No. 9Disease-modifying drugs for multiplesclerosis: a rapid and systematic review.By Clegg A, Bryant J, Milne R.No. 10Publication and related biases.A review by Song F, Eastwood AJ,Gilbody S, Duley L, Sutton AJ.No. 11Cost and outcome implications of theorganisation of vascular services.By Michaels J, Brazier J,Palfreyman S, Shackley P, Slack R.No. 12Monitoring blood glucose control indiabetes mellitus: a systematic review.By Coster S, Gulliford MC, Seed PT,Powrie JK, Swaminathan R.No. 13The effectiveness of domiciliaryhealth visiting: a systematic review ofinternational studies and a selectivereview of the British literature.By Elkan R, Kendrick D, Hewitt M,Robinson JJA, Tolley K, Blair M, et al.No. 14The determinants of screening uptakeand interventions for increasinguptake: a systematic review.By Jepson R, Clegg A, Forbes C,Lewis R, Sowden A, Kleijnen J.No. 15The effectiveness and cost-effectivenessof prophylactic removal of wisdomteeth.A rapid review by Song F, O’Meara S,Wilson P, Golder S, Kleijnen J.No. 16Ultrasound screening in pregnancy:a systematic review of the clinicaleffectiveness, cost-effectiveness andwomen’s views.By Bricker L, Garcia J, Henderson J,Mugford M, Neilson J, Roberts T, et al.No. 17A rapid and systematic review of theeffectiveness and cost-effectiveness ofthe taxanes used in the treatment ofadvanced breast and ovarian cancer.By Lister-Sharp D, McDonagh MS,Khan KS, Kleijnen J.No. 18Liquid-based cytology in cervicalscreening: a rapid and systematicreview.By Payne N, Chilcott J, McGoogan E.No. 19Randomised controlled trial of nondirectivecounselling, cognitive–behaviour therapy and usual generalpractitioner care in the management ofdepression as well as mixed anxiety anddepression in primary care.By King M, Sibbald B, Ward E,Bower P, Lloyd M, Gabbay M, et al.No. 20Routine referral for radiography ofpatients presenting with low back pain:is patients’ outcome influenced by GPs’referral for plain radiography?By Kerry S, Hilton S, Patel S,Dundas D, Rink E, Lord J.No. 21Systematic reviews of wound caremanagement: (3) antimicrobial agentsfor chronic wounds; (4) diabetic footulceration.By O’Meara S, Cullum N, Majid M,Sheldon T.No. 22Using routine data to complementand enhance the results of randomisedcontrolled trials.By Lewsey JD, Leyland AH, MurrayGD, Boddy FA.No. 23Coronary artery stents in the treatmentof ischaemic heart disease: a rapid andsystematic review.By Meads C, Cummins C, Jolly K,Stevens A, Burls A, Hyde C.No. 24Outcome measures for adult criticalcare: a systematic review.By Hayes JA, Black NA, Jenkinson C,Young JD, Rowan KM, Daly K, et al.No. 25A systematic review to evaluate theeffectiveness of interventions topromote the initiation of breastfeeding.By Fairbank L, O’Meara S,Renfrew MJ, Woolridge M, Sowden AJ,Lister-Sharp D.No. 26Implantable cardioverter defibrillators:arrhythmias. A rapid and systematicreview.By Parkes J, Bryant J, Milne R.No. 27Treatments for fatigue in multiplesclerosis: a rapid and systematic review.By Brañas P, Jordan R, Fry-Smith A,Burls A, Hyde C.No. 28Early asthma prophylaxis, naturalhistory, skeletal development andeconomy (EASE): a pilot randomisedcontrolled trial.By Baxter-Jones ADG, Helms PJ,Russell G, Grant A, Ross S, Cairns JA,et al.No. 29Screening for hypercholesterolaemiaversus case finding for familialhypercholesterolaemia: a systematicreview and cost-effectiveness analysis.By Marks D, WonderlingD, Thorogood M, Lambert H,Humphries SE, Neil HAW.No. 30A rapid and systematic review ofthe clinical effectiveness and costeffectivenessof glycoprotein IIb/IIIaantagonists in the medical managementof unstable angina.By McDonagh MS, Bachmann LM,Golder S, Kleijnen J, ter Riet G.335© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Health Technology Assessment reports published to date336No. 31A randomised controlled trialof prehospital intravenous fluidreplacement therapy in serious trauma.By Turner J, Nicholl J, Webber L,Cox H, Dixon S, Yates D.No. 32Intrathecal pumps for giving opioids inchronic pain: a systematic review.By Williams JE, Louw G,Towlerton G.No. 33Combination therapy (interferonalfa and ribavirin) in the treatmentof chronic hepatitis C: a rapid andsystematic review.By Shepherd J, Waugh N,Hewitson P.No. 34A systematic review of comparisons ofeffect sizes derived from randomisedand non-randomised studies.By MacLehose RR, Reeves BC,Harvey IM, Sheldon TA, Russell IT,Black AMS.No. 35Intravascular ultrasound-guidedinterventions in coronary arterydisease: a systematic literature review,with decision-analytic modelling, ofoutcomes and cost-effectiveness.By Berry E, Kelly S, Hutton J,Lindsay HSJ, Blaxill JM, Evans JA, et al.No. 36A randomised controlled trial toevaluate the effectiveness and costeffectivenessof counselling patientswith chronic depression.By Simpson S, Corney R,Fitzgerald P, Beecham J.No. 37Systematic review of treatments foratopic eczema.By Hoare C, Li Wan Po A,Williams H.No. 38Bayesian methods in health technologyassessment: a review.By Spiegelhalter DJ, Myles JP,Jones DR, Abrams KR.No. 39The management of dyspepsia: asystematic review.By Delaney B, Moayyedi P, Deeks J,Innes M, Soo S, Barton P, et al.No. 40A systematic review of treatments forsevere psoriasis.By Griffiths CEM, Clark CM,Chalmers RJG, Li Wan Po A,Williams HC.Volume 5, 2001No. 1Clinical and cost-effectivenessof donepezil, rivastigmine andgalantamine for Alzheimer’s disease: arapid and systematic review.By Clegg A, Bryant J, Nicholson T,McIntyre L, De Broe S, Gerard K, et al.No. 2The clinical effectiveness and costeffectivenessof riluzole for motorneurone disease: a rapid and systematicreview.By Stewart A, Sandercock J, Bryan S,Hyde C, Barton PM, Fry-Smith A, et al.No. 3Equity and the economic evaluation ofhealthcare.By Sassi F, Archard L, Le Grand J.No. 4Quality-of-life measures in chronicdiseases of childhood.By Eiser C, Morse R.No. 5Eliciting public preferences forhealthcare: a systematic review oftechniques.By Ryan M, Scott DA, Reeves C, BateA, van Teijlingen ER, Russell EM, et al.No. 6General health status measures forpeople with cognitive impairment:learning disability and acquired braininjury.By Riemsma RP, Forbes CA,Glanville JM, Eastwood AJ, Kleijnen J.No. 7An assessment of screening strategiesfor fragile X syndrome in the UK.By Pembrey ME, Barnicoat AJ,Carmichael B, Bobrow M, Turner G.No. 8Issues in methodological research:perspectives from researchers andcommissioners.By Lilford RJ, Richardson A, StevensA, Fitzpatrick R, Edwards S, Rock F, et al.No. 9Systematic reviews of woundcare management: (5) beds;(6) compression; (7) laser therapy,therapeutic ultrasound, electrotherapyand electromagnetic therapy.By Cullum N, Nelson EA,Flemming K, Sheldon T.No. 10Effects of educational and psychosocialinterventions for adolescents withdiabetes mellitus: a systematic review.By Hampson SE, Skinner TC, Hart J,Storey L, Gage H, Foxcroft D, et al.No. 11Effectiveness of autologous chondrocytetransplantation for hyaline cartilagedefects in knees: a rapid and systematicreview.By Jobanputra P, Parry D, Fry-SmithA, Burls A.No. 12Statistical assessment of the learningcurves of health technologies.By Ramsay CR, Grant AM, WallaceSA, Garthwaite PH, Monk AF, Russell IT.No. 13The effectiveness and cost-effectivenessof temozolomide for the treatment ofrecurrent malignant glioma: a rapidand systematic review.By Dinnes J, Cave C, Huang S,Major K, Milne R.No. 14A rapid and systematic review ofthe clinical effectiveness and costeffectivenessof debriding agents intreating surgical wounds healing bysecondary intention.By Lewis R, Whiting P, ter Riet G,O’Meara S, Glanville J.No. 15Home treatment for mental healthproblems: a systematic review.By Burns T, Knapp M, Catty J,Healey A, Henderson J, Watt H, et al.No. 16How to develop cost-consciousguidelines.By Eccles M, Mason J.No. 17The role of specialist nurses in multiplesclerosis: a rapid and systematic review.By De Broe S, Christopher F,Waugh N.No. 18A rapid and systematic reviewof the clinical effectiveness andcost-effectiveness of orlistat in themanagement of obesity.By O’Meara S, Riemsma R,Shirran L, Mather L, ter Riet G.No. 19The clinical effectiveness and costeffectivenessof pioglitazone fortype 2 diabetes mellitus: a rapid andsystematic review.By Chilcott J, Wight J, Lloyd JonesM, Tappenden P.No. 20Extended scope of nursing practice:a multicentre randomised controlledtrial of appropriately trained nursesand preregistration house officers inpreoperative assessment in electivegeneral surgery.By Kinley H, Czoski-Murray C,George S, McCabe C, Primrose J,Reilly C, et al.

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4No. 21Systematic reviews of the effectivenessof day care for people with severemental disorders: (1) Acute day hospitalversus admission; (2) Vocationalrehabilitation; (3) Day hospital versusoutpatient care.By Marshall M, Crowther R,Almaraz- Serrano A, Creed F, Sledge W,Kluiter H, et al.No. 22The measurement and monitoring ofsurgical adverse events.By Bruce J, Russell EM, Mollison J,Krukowski ZH.No. 31Design and use of questionnaires: areview of best practice applicable tosurveys of health service staff andpatients.By McColl E, Jacoby A, Thomas L,Soutter J, Bamford C, Steen N, et al.No. 32A rapid and systematic review ofthe clinical effectiveness and costeffectivenessof paclitaxel, docetaxel,gemcitabine and vinorelbine in nonsmall-celllung cancer.By Clegg A, Scott DA, Sidhu M,Hewitson P, Waugh N.No. 5The clinical effectiveness and costeffectivenessof inhaler devices usedin the routine management of chronicasthma in older children: a systematicreview and economic evaluation.By Peters J, Stevenson M, Beverley C,Lim J, Smith S.No. 6The clinical effectiveness and costeffectivenessof sibutramine in themanagement of obesity: a technologyassessment.By O’Meara S, Riemsma R, ShirranL, Mather L, ter Riet G.No. 23Action research: a systematic review andguidance for assessment.By Waterman H, Tillen D, Dickson R,de Koning K.No. 24A rapid and systematic review ofthe clinical effectiveness and costeffectivenessof gemcitabine for thetreatment of pancreatic cancer.By Ward S, Morris E, Bansback N,Calvert N, Crellin A, Forman D, et al.No. 25A rapid and systematic review of theevidence for the clinical effectivenessand cost-effectiveness of irinotecan,oxaliplatin and raltitrexed for thetreatment of advanced colorectalcancer.By Lloyd Jones M, Hummel S,Bansback N, Orr B, Seymour M.No. 26Comparison of the effectiveness ofinhaler devices in asthma and chronicobstructive airways disease: a systematicreview of the literature.By Brocklebank D, Ram F, Wright J,Barry P, Cates C, Davies L, et al.No. 27The cost-effectiveness of magneticresonance imaging for investigation ofthe knee joint.By Bryan S, Weatherburn G, BungayH, Hatrick C, Salas C, Parry D, et al.No. 28A rapid and systematic review ofthe clinical effectiveness and costeffectivenessof topotecan for ovariancancer.By Forbes C, Shirran L, Bagnall A-M,Duffy S, ter Riet G.No. 29Superseded by a report published in alater volume.No. 30The role of radiography in primarycare patients with low back pain of atleast 6 weeks duration: a randomised(unblinded) controlled trial.By Kendrick D, Fielding K, BentleyE, Miller P, Kerslake R, Pringle M.No. 33Subgroup analyses in randomisedcontrolled trials: quantifying the risksof false-positives and false-negatives.By Brookes ST, Whitley E, Peters TJ,Mulheran PA, Egger M, Davey Smith G.No. 34Depot antipsychotic medicationin the treatment of patients withschizophrenia: (1) Meta-review; (2)Patient and nurse attitudes.By David AS, Adams C.No. 35A systematic review of controlledtrials of the effectiveness and costeffectivenessof brief psychologicaltreatments for depression.By Churchill R, Hunot V, Corney R,Knapp M, McGuire H, Tylee A, et al.No. 36Cost analysis of child healthsurveillance.By Sanderson D, Wright D, Acton C,Duree D.Volume 6, 2002No. 1A study of the methods used to selectreview criteria for clinical audit.By Hearnshaw H, Harker R,Cheater F, Baker R, Grimshaw G.No. 2Fludarabine as second-line therapy forB cell chronic lymphocytic leukaemia: atechnology assessment.By Hyde C, Wake B, Bryan S, BartonP, Fry-Smith A, Davenport C, et al.No. 3Rituximab as third-line treatment forrefractory or recurrent Stage III or IVfollicular non-Hodgkin’s lymphoma:a systematic review and economicevaluation.By Wake B, Hyde C, Bryan S, BartonP, Song F, Fry-Smith A, et al.No. 4A systematic review of dischargearrangements for older people.By Parker SG, Peet SM, McPhersonA, Cannaby AM, Baker R, Wilson A, et al.No. 7The cost-effectiveness of magneticresonance angiography for carotidartery stenosis and peripheral vasculardisease: a systematic review.By Berry E, Kelly S, Westwood ME,Davies LM, Gough MJ, Bamford JM,et al.No. 8Promoting physical activity in SouthAsian Muslim women through ‘exerciseon prescription’.By Carroll B, Ali N, Azam N.No. 9Zanamivir for the treatment ofinfluenza in adults: a systematic reviewand economic evaluation.By Burls A, Clark W, Stewart T,Preston C, Bryan S, Jefferson T, et al.No. 10A review of the natural history andepidemiology of multiple sclerosis:implications for resource allocation andhealth economic models.By Richards RG, Sampson FC,Beard SM, Tappenden P.No. 11Screening for gestational diabetes:a systematic review and economicevaluation.By Scott DA, Loveman E, McIntyreL, Waugh N.No. 12The clinical effectiveness and costeffectivenessof surgery for people withmorbid obesity: a systematic review andeconomic evaluation.By Clegg AJ, Colquitt J, Sidhu MK,Royle P, Loveman E, Walker A.No. 13The clinical effectiveness oftrastuzumab for breast cancer: asystematic review.By Lewis R, Bagnall A-M, Forbes C,Shirran E, Duffy S, Kleijnen J, et al.No. 14The clinical effectiveness and costeffectivenessof vinorelbine for breastcancer: a systematic review andeconomic evaluation.By Lewis R, Bagnall A-M, King S,Woolacott N, Forbes C, Shirran L, et al.337© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Health Technology Assessment reports published to dateNo. 15A systematic review of the effectivenessand cost-effectiveness of metal-onmetalhip resurfacing arthroplasty fortreatment of hip disease.By Vale L, Wyness L, McCormack K,McKenzie L, Brazzelli M, Stearns SC.No. 16The clinical effectiveness and costeffectivenessof bupropion and nicotinereplacement therapy for smokingcessation: a systematic review andeconomic evaluation.By Woolacott NF, Jones L, Forbes CA,Mather LC, Sowden AJ, Song FJ, et al.No. 17A systematic review of effectivenessand economic evaluation of new drugtreatments for juvenile idiopathicarthritis: etanercept.By Cummins C, Connock M,Fry-Smith A, Burls A.No. 24A systematic review of the effectivenessof interventions based on a stages-ofchangeapproach to promote individualbehaviour change.By Riemsma RP, Pattenden J, BridleC, Sowden AJ, Mather L, Watt IS, et al.No. 25A systematic review update of theclinical effectiveness and costeffectivenessof glycoprotein IIb/IIIaantagonists.By Robinson M, Ginnelly L, SculpherM, Jones L, Riemsma R, Palmer S, et al.No. 26A systematic review of the effectiveness,cost-effectiveness and barriers toimplementation of thrombolytic andneuroprotective therapy for acuteischaemic stroke in the NHS.By Sandercock P, Berge E, Dennis M,Forbes J, Hand P, Kwan J, et al.No. 33The effectiveness and cost-effectivenessof imatinib in chronic myeloidleukaemia: a systematic review.By Garside R, Round A, Dalziel K,Stein K, Royle R.No. 34A comparative study of hypertonicsaline, daily and alternate-day rhDNasein children with cystic fibrosis.By Suri R, Wallis C, Bush A,Thompson S, Normand C, Flather M,et al.No. 35A systematic review of the costs andeffectiveness of different models ofpaediatric home care.By Parker G, Bhakta P, Lovett CA,Paisley S, Olsen R, Turner D, et al.Volume 7, 2003338No. 18Clinical effectiveness and costeffectivenessof growth hormone inchildren: a systematic review andeconomic evaluation.By Bryant J, Cave C, Mihaylova B,Chase D, McIntyre L, Gerard K, et al.No. 19Clinical effectiveness and costeffectivenessof growth hormonein adults in relation to impact onquality of life: a systematic review andeconomic evaluation.By Bryant J, Loveman E, Chase D,Mihaylova B, Cave C, Gerard K, et al.No. 20Clinical medication review by apharmacist of patients on repeatprescriptions in general practice: arandomised controlled trial.By Zermansky AG, Petty DR, RaynorDK, Lowe CJ, Freementle N, Vail A.No. 21The effectiveness of infliximab andetanercept for the treatment ofrheumatoid arthritis: a systematicreview and economic evaluation.By Jobanputra P, Barton P, Bryan S,Burls A.No. 22A systematic review and economicevaluation of computerised cognitivebehaviour therapy for depression andanxiety.By Kaltenthaler E, Shackley P,Stevens K, Beverley C, Parry G,Chilcott J.No. 23A systematic review and economicevaluation of pegylated liposomaldoxorubicin hydrochloride for ovariancancer.By Forbes C, Wilby J, Richardson G,Sculpher M, Mather L, Reimsma R.No. 27A randomised controlled crossover trialof nurse practitioner versus doctorledoutpatient care in a bronchiectasisclinic.By Caine N, Sharples LD,Hollingworth W, French J, Keogan M,Exley A, et al.No. 28Clinical effectiveness and cost –consequences of selective serotoninreuptake inhibitors in the treatment ofsex offenders.By Adi Y, Ashcroft D, Browne K,Beech A, Fry-Smith A, Hyde C.No. 29Treatment of established osteoporosis:a systematic review and cost–utilityanalysis.By Kanis JA, Brazier JE, StevensonM, Calvert NW, Lloyd Jones M.No. 30Which anaesthetic agents are costeffectivein day surgery? Literaturereview, national survey of practice andrandomised controlled trial.By Elliott RA Payne K, Moore JK,Davies LM, Harper NJN, St Leger AS,et al.No. 31Screening for hepatitis C amonginjecting drug users and ingenitourinary medicine clinics:systematic reviews of effectiveness,modelling study and national survey ofcurrent practice.By Stein K, Dalziel K, Walker A,McIntyre L, Jenkins B, Horne J, et al.No. 32The measurement of satisfaction withhealthcare: implications for practicefrom a systematic review of theliterature.By Crow R, Gage H, Hampson S,Hart J, Kimber A, Storey L, et al.No. 1How important are comprehensiveliterature searches and the assessmentof trial quality in systematic reviews?Empirical study.By Egger M, Jüni P, Bartlett C,Holenstein F, Sterne J.No. 2Systematic review of the effectivenessand cost-effectiveness, and economicevaluation, of home versus hospital orsatellite unit haemodialysis for peoplewith end-stage renal failure.By Mowatt G, Vale L, Perez J, WynessL, Fraser C, MacLeod A, et al.No. 3Systematic review and economicevaluation of the effectiveness ofinfliximab for the treatment of Crohn’sdisease.By Clark W, Raftery J, Barton P,Song F, Fry-Smith A, Burls A.No. 4A review of the clinical effectivenessand cost-effectiveness of routine anti-Dprophylaxis for pregnant women whoare rhesus negative.By Chilcott J, Lloyd Jones M, WightJ, Forman K, Wray J, Beverley C, et al.No. 5Systematic review and evaluation of theuse of tumour markers in paediatriconcology: Ewing’s sarcoma andneuroblastoma.By Riley RD, Burchill SA,Abrams KR, Heney D, Lambert PC,Jones DR, et al.No. 6The cost-effectiveness of screening forHelicobacter pylori to reduce mortalityand morbidity from gastric cancer andpeptic ulcer disease: a discrete-eventsimulation model.By Roderick P, Davies R, Raftery J,Crabbe D, Pearce R, Bhandari P, et al.

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4No. 7The clinical effectiveness and costeffectivenessof routine dental checks:a systematic review and economicevaluation.By Davenport C, Elley K, SalasC, Taylor-Weetman CL, Fry-Smith A,Bryan S, et al.No. 8A multicentre randomised controlledtrial assessing the costs and benefitsof using structured information andanalysis of women’s preferences in themanagement of menorrhagia.By Kennedy ADM, Sculpher MJ,Coulter A, Dwyer N, Rees M, Horsley S,et al.No. 9Clinical effectiveness and cost–utilityof photodynamic therapy for wetage-related macular degeneration:a systematic review and economicevaluation.By Meads C, Salas C, Roberts T,Moore D, Fry-Smith A, Hyde C.No. 10Evaluation of molecular tests forprenatal diagnosis of chromosomeabnormalities.By Grimshaw GM, Szczepura A,Hultén M, MacDonald F, Nevin NC,Sutton F, et al.No. 11First and second trimester antenatalscreening for Down’s syndrome:the results of the Serum, Urine andUltrasound Screening Study (SURUSS).By Wald NJ, Rodeck C, HackshawAK, Walters J, Chitty L, Mackinson AM.No. 12The effectiveness and cost-effectivenessof ultrasound locating devices forcentral venous access: a systematicreview and economic evaluation.By Calvert N, Hind D, McWilliamsRG, Thomas SM, Beverley C,Davidson A.No. 13A systematic review of atypicalantipsychotics in schizophrenia.By Bagnall A-M, Jones L, Lewis R,Ginnelly L, Glanville J, Torgerson D,et al.No. 14Prostate Testing for Cancer andTreatment (ProtecT) feasibility study.By Donovan J, Hamdy F, Neal D,Peters T, Oliver S, Brindle L, et al.No. 15Early thrombolysis for the treatmentof acute myocardial infarction: asystematic review and economicevaluation.By Boland A, Dundar Y, Bagust A,Haycox A, Hill R, Mujica Mota R, et al.No. 16Screening for fragile X syndrome: aliterature review and modelling.By Song FJ, Barton P, SleightholmeV, Yao GL, Fry-Smith A.No. 17Systematic review of endoscopic sinussurgery for nasal polyps.By Dalziel K, Stein K, Round A,Garside R, Royle P.No. 18Towards efficient guidelines: how tomonitor guideline use in primary care.By Hutchinson A, McIntosh A,Cox S, Gilbert C.No. 19Effectiveness and cost-effectivenessof acute hospital-based spinal cordinjuries services: systematic review.By Bagnall A-M, Jones L, RichardsonG, Duffy S, Riemsma R.No. 20Prioritisation of health technologyassessment. The PATHS model:methods and case studies.By Townsend J, Buxton M,Harper G.No. 21Systematic review of the clinicaleffectiveness and cost-effectiveness oftension-free vaginal tape for treatmentof urinary stress incontinence.By Cody J, Wyness L, Wallace S,Glazener C, Kilonzo M, Stearns S, et al.No. 22The clinical and cost-effectiveness ofpatient education models for diabetes:a systematic review and economicevaluation.By Loveman E, Cave C, Green C,Royle P, Dunn N, Waugh N.No. 23The role of modelling in prioritisingand planning clinical trials.By Chilcott J, Brennan A, Booth A,Karnon J, Tappenden P.No. 24Cost–benefit evaluation of routineinfluenza immunisation in people65–74 years of age.By Allsup S, Gosney M, Haycox A,Regan M.No. 25The clinical and cost-effectiveness ofpulsatile machine perfusion versus coldstorage of kidneys for transplantationretrieved from heart-beating and nonheart-beatingdonors.By Wight J, Chilcott J, Holmes M,Brewer N.No. 26Can randomised trials rely on existingelectronic data? A feasibility study toexplore the value of routine data inhealth technology assessment.By Williams JG, Cheung WY,Cohen DR, Hutchings HA, Longo MF,Russell IT.No. 27Evaluating non-randomisedintervention studies.By Deeks JJ, Dinnes J, D’Amico R,Sowden AJ, Sakarovitch C, Song F, et al.No. 28A randomised controlled trial to assessthe impact of a package comprising apatient-orientated, evidence-based selfhelpguidebook and patient-centredconsultations on disease managementand satisfaction in inflammatory boweldisease.By Kennedy A, Nelson E, Reeves D,Richardson G, Roberts C, Robinson A,et al.No. 29The effectiveness of diagnostic tests forthe assessment of shoulder pain dueto soft tissue disorders: a systematicreview.By Dinnes J, Loveman E, McIntyre L,Waugh N.No. 30The value of digital imaging in diabeticretinopathy.By Sharp PF, Olson J, Strachan F,Hipwell J, Ludbrook A, O’Donnell M,et al.No. 31Lowering blood pressure to preventmyocardial infarction and stroke: a newpreventive strategy.By Law M, Wald N, Morris J.No. 32Clinical and cost-effectiveness ofcapecitabine and tegafur with uracil forthe treatment of metastatic colorectalcancer: systematic review and economicevaluation.By Ward S, Kaltenthaler E, Cowan J,Brewer N.No. 33Clinical and cost-effectiveness of newand emerging technologies for earlylocalised prostate cancer: a systematicreview.By Hummel S, Paisley S, Morgan A,Currie E, Brewer N.No. 34Literature searching for clinical andcost-effectiveness studies used in healthtechnology assessment reports carriedout for the National Institute forClinical Excellence appraisal system.By Royle P, Waugh N.339© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Health Technology Assessment reports published to dateNo. 35Systematic review and economicdecision modelling for the preventionand treatment of influenza A and B.By Turner D, Wailoo A, Nicholson K,Cooper N, Sutton A, Abrams K.No. 36A randomised controlled trialto evaluate the clinical and costeffectivenessof Hickman line insertionsin adult cancer patients by nurses.By Boland A, Haycox A, Bagust A,Fitzsimmons L.No. 37Redesigning postnatal care: arandomised controlled trial of protocolbasedmidwifery-led care focusedon individual women’s physical andpsychological health needs.By MacArthur C, Winter HR,Bick DE, Lilford RJ, Lancashire RJ,Knowles H, et al.No. 38Estimating implied rates of discount inhealthcare decision-making.By West RR, McNabb R, ThompsonAGH, Sheldon TA, Grimley Evans J.No. 39Systematic review of isolation policiesin the hospital management ofmethicillin-resistant Staphylococcusaureus: a review of the literaturewith epidemiological and economicmodelling.By Cooper BS, Stone SP, Kibbler CC,Cookson BD, Roberts JA, Medley GF,et al.No. 40Treatments for spasticity and pain inmultiple sclerosis: a systematic review.By Beard S, Hunn A, Wight J.No. 41The inclusion of reports of randomisedtrials published in languages other thanEnglish in systematic reviews.By Moher D, Pham B, Lawson ML,Klassen TP.No. 2Systematic review and modelling of theinvestigation of acute and chronic chestpain presenting in primary care.By Mant J, McManus RJ, Oakes RAL,Delaney BC, Barton PM, Deeks JJ, et al.No. 3The effectiveness and cost-effectivenessof microwave and thermal balloonendometrial ablation for heavymenstrual bleeding: a systematic reviewand economic modelling.By Garside R, Stein K, Wyatt K,Round A, Price A.No. 4A systematic review of the role ofbisphosphonates in metastatic disease.By Ross JR, Saunders Y,Edmonds PM, Patel S, Wonderling D,Normand C, et al.No. 5Systematic review of the clinicaleffectiveness and cost-effectivenessof capecitabine (Xeloda ® ) for locallyadvanced and/or metastatic breastcancer.By Jones L, Hawkins N, Westwood M,Wright K, Richardson G, Riemsma R.No. 6Effectiveness and efficiency of guidelinedissemination and implementationstrategies.By Grimshaw JM, Thomas RE,MacLennan G, Fraser C, Ramsay CR,Vale L, et al.No. 7Clinical effectiveness and costs of theSugarbaker procedure for the treatmentof pseudomyxoma peritonei.By Bryant J, Clegg AJ, Sidhu MK,Brodin H, Royle P, Davidson P.No. 8Psychological treatment for insomniain the regulation of long-term hypnoticdrug use.By Morgan K, Dixon S, Mathers N,Thompson J, Tomeny M.No. 11The use of modelling to evaluatenew drugs for patients with a chroniccondition: the case of antibodiesagainst tumour necrosis factor inrheumatoid arthritis.By Barton P, Jobanputra P, Wilson J,Bryan S, Burls A.No. 12Clinical effectiveness and costeffectivenessof neonatal screeningfor inborn errors of metabolism usingtandem mass spectrometry: a systematicreview.By Pandor A, Eastham J, Beverley C,Chilcott J, Paisley S.No. 13Clinical effectiveness and costeffectivenessof pioglitazone androsiglitazone in the treatment of type2 diabetes: a systematic review andeconomic evaluation.By Czoski-Murray C, Warren E,Chilcott J, Beverley C, Psyllaki MA,Cowan J.No. 14Routine examination of the newborn:the EMREN study. Evaluation of anextension of the midwife role includinga randomised controlled trial ofappropriately trained midwives andpaediatric senior house officers.By Townsend J, Wolke D, Hayes J,Davé S, Rogers C, Bloomfield L, et al.No. 15Involving consumers in research anddevelopment agenda setting for theNHS: developing an evidence-basedapproach.By Oliver S, Clarke-Jones L, Rees R,Milne R, Buchanan P, Gabbay J, et al.No. 16A multi-centre randomised controlledtrial of minimally invasive directcoronary bypass grafting versuspercutaneous transluminal coronaryangioplasty with stenting for proximalstenosis of the left anterior descendingcoronary artery.By Reeves BC, Angelini GD, BryanAJ, Taylor FC, Cripps T, Spyt TJ, et al.340No. 42The impact of screening on futurehealth-promoting behaviours andhealth beliefs: a systematic review.By Bankhead CR, Brett J, Bukach C,Webster P, Stewart-Brown S, Munafo M,et al.Volume 8, 2004No. 1What is the best imaging strategy foracute stroke?By Wardlaw JM, Keir SL, Seymour J,Lewis S, Sandercock PAG, Dennis MS,et al.No. 9Improving the evaluation oftherapeutic interventions in multiplesclerosis: development of a patientbasedmeasure of outcome.By Hobart JC, Riazi A, Lamping DL,Fitzpatrick R, Thompson AJ.No. 10A systematic review and economicevaluation of magnetic resonancecholangiopancreatography comparedwith diagnostic endoscopic retrogradecholangiopancreatography.By Kaltenthaler E, Bravo Vergel Y,Chilcott J, Thomas S, Blakeborough T,Walters SJ, et al.No. 17Does early magnetic resonance imaginginfluence management or improveoutcome in patients referred tosecondary care with low back pain? Apragmatic randomised controlled trial.By Gilbert FJ, Grant AM, GillanMGC, Vale L, Scott NW, Campbell MK,et al.No. 18The clinical and cost-effectivenessof anakinra for the treatment ofrheumatoid arthritis in adults: asystematic review and economicanalysis.By Clark W, Jobanputra P, Barton P,Burls A.

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4No. 19A rapid and systematic review andeconomic evaluation of the clinicaland cost-effectiveness of newer drugsfor treatment of mania associated withbipolar affective disorder.By Bridle C, Palmer S, Bagnall A-M,Darba J, Duffy S, Sculpher M, et al.No. 20Liquid-based cytology in cervicalscreening: an updated rapid andsystematic review and economicanalysis.By Karnon J, Peters J, Platt J,Chilcott J, McGoogan E, Brewer N.No. 21Systematic review of the long-termeffects and economic consequences oftreatments for obesity and implicationsfor health improvement.By Avenell A, Broom J, Brown TJ,Poobalan A, Aucott L, Stearns SC, et al.No. 22Autoantibody testing in childrenwith newly diagnosed type 1 diabetesmellitus.By Dretzke J, Cummins C,Sandercock J, Fry-Smith A, Barrett T,Burls A.No. 23Clinical effectiveness and costeffectivenessof prehospital intravenousfluids in trauma patients.By Dretzke J, Sandercock J, BaylissS, Burls A.No. 24Newer hypnotic drugs for the shorttermmanagement of insomnia: asystematic review and economicevaluation.By Dündar Y, Boland A, Strobl J,Dodd S, Haycox A, Bagust A, et al.No. 25Development and validation ofmethods for assessing the quality ofdiagnostic accuracy studies.By Whiting P, Rutjes AWS, Dinnes J,Reitsma JB, Bossuyt PMM, Kleijnen J.No. 26EVALUATE hysterectomy trial:a multicentre randomised trialcomparing abdominal, vaginal andlaparoscopic methods of hysterectomy.By Garry R, Fountain J, Brown J,Manca A, Mason S, Sculpher M, et al.No. 27Methods for expected value ofinformation analysis in complex healtheconomic models: developments onthe health economics of interferon-βand glatiramer acetate for multiplesclerosis.By Tappenden P, Chilcott JB,Eggington S, Oakley J, McCabe C.No. 28Effectiveness and cost-effectivenessof imatinib for first-line treatmentof chronic myeloid leukaemia inchronic phase: a systematic review andeconomic analysis.By Dalziel K, Round A, Stein K,Garside R, Price A.No. 29VenUS I: a randomised controlled trialof two types of bandage for treatingvenous leg ulcers.By Iglesias C, Nelson EA, CullumNA, Torgerson DJ, on behalf of theVenUS Team.No. 30Systematic review of the effectivenessand cost-effectiveness, and economicevaluation, of myocardial perfusionscintigraphy for the diagnosis andmanagement of angina and myocardialinfarction.By Mowatt G, Vale L, Brazzelli M,Hernandez R, Murray A, Scott N, et al.No. 31A pilot study on the use of decisiontheory and value of informationanalysis as part of the NHS HealthTechnology Assessment programme.By Claxton K, Ginnelly L, SculpherM, Philips Z, Palmer S.No. 32The Social Support and Family HealthStudy: a randomised controlled trialand economic evaluation of twoalternative forms of postnatal supportfor mothers living in disadvantagedinner-city areas.By Wiggins M, Oakley A, Roberts I,Turner H, Rajan L, Austerberry H, et al.No. 33Psychosocial aspects of geneticscreening of pregnant women andnewborns: a systematic review.By Green JM, Hewison J, Bekker HL,Bryant, Cuckle HS.No. 34Evaluation of abnormal uterinebleeding: comparison of threeoutpatient procedures within cohortsdefined by age and menopausal status.By Critchley HOD, Warner P, Lee AJ,Brechin S, Guise J, Graham B.No. 35Coronary artery stents: a rapidsystematic review and economicevaluation.By Hill R, Bagust A, Bakhai A,Dickson R, Dündar Y, Haycox A, et al.No. 36Review of guidelines for good practicein decision-analytic modelling in healthtechnology assessment.By Philips Z, Ginnelly L, Sculpher M,Claxton K, Golder S, Riemsma R, et al.No. 37Rituximab (MabThera ® ) for aggressivenon-Hodgkin’s lymphoma: systematicreview and economic evaluation.By Knight C, Hind D, Brewer N,Abbott V.No. 38Clinical effectiveness and costeffectivenessof clopidogrel andmodified-release dipyridamole in thesecondary prevention of occlusivevascular events: a systematic review andeconomic evaluation.By Jones L, Griffin S, Palmer S, MainC, Orton V, Sculpher M, et al.No. 39Pegylated interferon α-2a and -2bin combination with ribavirin in thetreatment of chronic hepatitis C:a systematic review and economicevaluation.By Shepherd J, Brodin H, Cave C,Waugh N, Price A, Gabbay J.No. 40Clopidogrel used in combination withaspirin compared with aspirin alonein the treatment of non-ST-segmentelevationacute coronary syndromes:a systematic review and economicevaluation.By Main C, Palmer S, Griffin S, JonesL, Orton V, Sculpher M, et al.No. 41Provision, uptake and cost of cardiacrehabilitation programmes: improvingservices to under-represented groups.By Beswick AD, Rees K, Griebsch I,Taylor FC, Burke M, West RR, et al.No. 42Involving South Asian patients inclinical trials.By Hussain-Gambles M, Leese B,Atkin K, Brown J, Mason S, Tovey P.No. 43Clinical and cost-effectiveness ofcontinuous subcutaneous insulininfusion for diabetes.By Colquitt JL, Green C, Sidhu MK,Hartwell D, Waugh N.No. 44Identification and assessment ofongoing trials in health technologyassessment reviews.By Song FJ, Fry-Smith A, DavenportC, Bayliss S, Adi Y, Wilson JS, et al.No. 45Systematic review and economicevaluation of a long-acting insulinanalogue, insulin glargineBy Warren E, Weatherley-Jones E,Chilcott J, Beverley C.341© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Health Technology Assessment reports published to date342No. 46Supplementation of a home-basedexercise programme with a classbasedprogramme for peoplewith osteoarthritis of the knees: arandomised controlled trial and healtheconomic analysis.By McCarthy CJ, Mills PM, Pullen R,Richardson G, Hawkins N, Roberts CR,et al.No. 47Clinical and cost-effectiveness of oncedailyversus more frequent use of samepotency topical corticosteroids foratopic eczema: a systematic review andeconomic evaluation.By Green C, Colquitt JL, Kirby J,Davidson P, Payne E.No. 48Acupuncture of chronic headachedisorders in primary care: randomisedcontrolled trial and economic analysis.By Vickers AJ, Rees RW, Zollman CE,McCarney R, Smith CM, Ellis N, et al.No. 49Generalisability in economic evaluationstudies in healthcare: a review and casestudies.By Sculpher MJ, Pang FS, Manca A,Drummond MF, Golder S, Urdahl H,et al.No. 50Virtual outreach: a randomisedcontrolled trial and economicevaluation of joint teleconferencedmedical consultations.By Wallace P, Barber J, Clayton W,Currell R, Fleming K, Garner P, et al.Volume 9, 2005No. 1Randomised controlled multipletreatment comparison to provide a costeffectivenessrationale for the selectionof antimicrobial therapy in acne.By Ozolins M, Eady EA, Avery A,Cunliffe WJ, O’Neill C, Simpson NB,et al.No. 2Do the findings of case series studiesvary significantly according tomethodological characteristics?By Dalziel K, Round A, Stein K,Garside R, Castelnuovo E, Payne L.No. 3Improving the referral processfor familial breast cancer geneticcounselling: findings of threerandomised controlled trials of twointerventions.By Wilson BJ, Torrance N,Mollison J, Wordsworth S, Gray JR,Haites NE, et al.No. 4Randomised evaluation of alternativeelectrosurgical modalities to treatbladder outflow obstruction in menwith benign prostatic hyperplasia.By Fowler C, McAllister W, Plail R,Karim O, Yang Q.No. 5A pragmatic randomised controlledtrial of the cost-effectiveness ofpalliative therapies for patients withinoperable oesophageal cancer.By Shenfine J, McNamee P, Steen N,Bond J, Griffin SM.No. 6Impact of computer-aided detectionprompts on the sensitivity andspecificity of screening mammography.By Taylor P, Champness J, Given-Wilson R, Johnston K, Potts H.No. 7Issues in data monitoring and interimanalysis of trials.By Grant AM, Altman DG, BabikerAB, Campbell MK, Clemens FJ,Darbyshire JH, et al.No. 8Lay public’s understanding of equipoiseand randomisation in randomisedcontrolled trials.By Robinson EJ, Kerr CEP,Stevens AJ, Lilford RJ, Braunholtz DA,Edwards SJ, et al.No. 9Clinical and cost-effectiveness ofelectroconvulsive therapy for depressiveillness, schizophrenia, catatoniaand mania: systematic reviews andeconomic modelling studies.By Greenhalgh J, Knight C, Hind D,Beverley C, Walters S.No. 10Measurement of health-related qualityof life for people with dementia:development of a new instrument(DEMQOL) and an evaluation ofcurrent methodology.By Smith SC, Lamping DL, BanerjeeS, Harwood R, Foley B, Smith P, et al.No. 11Clinical effectiveness and costeffectivenessof drotrecogin alfa(activated) (Xigris ® ) for the treatmentof severe sepsis in adults: a systematicreview and economic evaluation.By Green C, Dinnes J, Takeda A,Shepherd J, Hartwell D, Cave C, et al.No. 12A methodological review of howheterogeneity has been examined insystematic reviews of diagnostic testaccuracy.By Dinnes J, Deeks J, Kirby J,Roderick P.No. 13Cervical screening programmes: canautomation help? Evidence fromsystematic reviews, an economicanalysis and a simulation modellingexercise applied to the UK.By Willis BH, Barton P, Pearmain P,Bryan S, Hyde C.No. 14Laparoscopic surgery for inguinalhernia repair: systematic review ofeffectiveness and economic evaluation.By McCormack K, Wake B, Perez J,Fraser C, Cook J, McIntosh E, et al.No. 15Clinical effectiveness, tolerability andcost-effectiveness of newer drugs forepilepsy in adults: a systematic reviewand economic evaluation.By Wilby J, Kainth A, Hawkins N,Epstein D, McIntosh H, McDaid C, et al.No. 16A randomised controlled trial tocompare the cost-effectiveness oftricyclic antidepressants, selectiveserotonin reuptake inhibitors andlofepramine.By Peveler R, Kendrick T, Buxton M,Longworth L, Baldwin D, Moore M, et al.No. 17Clinical effectiveness and costeffectivenessof immediate angioplastyfor acute myocardial infarction:systematic review and economicevaluation.By Hartwell D, Colquitt J, LovemanE, Clegg AJ, Brodin H, Waugh N, et al.No. 18A randomised controlled comparison ofalternative strategies in stroke care.By Kalra L, Evans A, Perez I,Knapp M, Swift C, Donaldson N.No. 19The investigation and analysis ofcritical incidents and adverse events inhealthcare.By Woloshynowych M, Rogers S,Taylor-Adams S, Vincent C.No. 20Potential use of routine databases inhealth technology assessment.By Raftery J, Roderick P, Stevens A.No. 21Clinical and cost-effectiveness of newerimmunosuppressive regimens in renaltransplantation: a systematic review andmodelling study.By Woodroffe R, Yao GL, Meads C,Bayliss S, Ready A, Raftery J, et al.No. 22A systematic review and economicevaluation of alendronate, etidronate,risedronate, raloxifene and teriparatidefor the prevention and treatment ofpostmenopausal osteoporosis.By Stevenson M, Lloyd Jones M, DeNigris E, Brewer N, Davis S, Oakley J.

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4No. 23A systematic review to examinethe impact of psycho-educationalinterventions on health outcomesand costs in adults and children withdifficult asthma.By Smith JR, Mugford M, HollandR, Candy B, Noble MJ, Harrison BDW,et al.No. 24An evaluation of the costs, effectivenessand quality of renal replacementtherapy provision in renal satellite unitsin England and Wales.By Roderick P, Nicholson T, ArmitageA, Mehta R, Mullee M, Gerard K, et al.No. 25Imatinib for the treatment of patientswith unresectable and/or metastaticgastrointestinal stromal tumours:systematic review and economicevaluation.By Wilson J, Connock M, Song F,Yao G, Fry-Smith A, Raftery J, et al.No. 26Indirect comparisons of competinginterventions.By Glenny AM, Altman DG, Song F,Sakarovitch C, Deeks JJ, D’Amico R,et al.No. 27Cost-effectiveness of alternativestrategies for the initial medicalmanagement of non-ST elevation acutecoronary syndrome: systematic reviewand decision-analytical modelling.By Robinson M, Palmer S, SculpherM, Philips Z, Ginnelly L, Bowens A, et al.No. 28Outcomes of electrically stimulatedgracilis neosphincter surgery.By Tillin T, Chambers M, Feldman R.No. 29The effectiveness and cost-effectivenessof pimecrolimus and tacrolimus foratopic eczema: a systematic review andeconomic evaluation.By Garside R, Stein K, CastelnuovoE, Pitt M, Ashcroft D, Dimmock P, et al.No. 30Systematic review on urine albumintesting for early detection of diabeticcomplications.By Newman DJ, Mattock MB,Dawnay ABS, Kerry S, McGuire A,Yaqoob M, et al.No. 31Randomised controlled trial of the costeffectivenessof water-based therapy forlower limb osteoarthritis.By Cochrane T, Davey RC,Matthes Edwards SM.No. 32Longer term clinical and economicbenefits of offering acupuncture care topatients with chronic low back pain.By Thomas KJ, MacPhersonH, Ratcliffe J, Thorpe L, Brazier J,Campbell M, et al.No. 33Cost-effectiveness and safety ofepidural steroids in the managementof sciatica.By Price C, Arden N, Coglan L,Rogers P.No. 34The British Rheumatoid OutcomeStudy Group (BROSG) randomisedcontrolled trial to compare theeffectiveness and cost-effectiveness ofaggressive versus symptomatic therapyin established rheumatoid arthritis.By Symmons D, Tricker K, RobertsC, Davies L, Dawes P, Scott DL.No. 35Conceptual framework and systematicreview of the effects of participants’and professionals’ preferences inrandomised controlled trials.By King M, Nazareth I, Lampe F,Bower P, Chandler M, Morou M, et al.No. 36The clinical and cost-effectiveness ofimplantable cardioverter defibrillators:a systematic review.By Bryant J, Brodin H, Loveman E,Payne E, Clegg A.No. 37A trial of problem-solving bycommunity mental health nurses foranxiety, depression and life difficultiesamong general practice patients. TheCPN-GP study.By Kendrick T, Simons L,Mynors-Wallis L, Gray A, Lathlean J,Pickering R, et al.No. 38The causes and effects of sociodemographicexclusions from clinicaltrials.By Bartlett C, Doyal L, Ebrahim S,Davey P, Bachmann M, Egger M, et al.No. 39Is hydrotherapy cost-effective?A randomised controlled trial ofcombined hydrotherapy programmescompared with physiotherapy landtechniques in children with juvenileidiopathic arthritis.By Epps H, Ginnelly L, Utley M,Southwood T, Gallivan S, Sculpher M,et al.No. 40A randomised controlled trial andcost-effectiveness study of systematicscreening (targeted and totalpopulation screening) versus routinepractice for the detection of atrialfibrillation in people aged 65 and over.The SAFE study.By Hobbs FDR, Fitzmaurice DA,Mant J, Murray E, Jowett S, Bryan S,et al.No. 41Displaced intracapsular hip fracturesin fit, older people: a randomisedcomparison of reduction and fixation,bipolar hemiarthroplasty and total hiparthroplasty.By Keating JF, Grant A, Masson M,Scott NW, Forbes JF.No. 42Long-term outcome of cognitivebehaviour therapy clinical trials incentral Scotland.By Durham RC, Chambers JA,Power KG, Sharp DM, Macdonald RR,Major KA, et al.No. 43The effectiveness and cost-effectivenessof dual-chamber pacemakers comparedwith single-chamber pacemakers forbradycardia due to atrioventricularblock or sick sinus syndrome: systematicreview and economic evaluation.By Castelnuovo E, Stein K, Pitt M,Garside R, Payne E.No. 44Newborn screening for congenital heartdefects: a systematic review and costeffectivenessanalysis.By Knowles R, Griebsch I,Dezateux C, Brown J, Bull C, Wren C.No. 45The clinical and cost-effectiveness ofleft ventricular assist devices for endstageheart failure: a systematic reviewand economic evaluation.By Clegg AJ, Scott DA, Loveman E,Colquitt J, Hutchinson J, Royle P, et al.No. 46The effectiveness of the HeidelbergRetina Tomograph and laser diagnosticglaucoma scanning system (GDx) indetecting and monitoring glaucoma.By Kwartz AJ, Henson DB, HarperRA, Spencer AF, McLeod D.No. 47Clinical and cost-effectiveness ofautologous chondrocyte implantationfor cartilage defects in knee joints:systematic review and economicevaluation.By Clar C, Cummins E, McIntyre L,Thomas S, Lamb J, Bain L, et al.343© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Health Technology Assessment reports published to date344No. 48Systematic review of effectiveness ofdifferent treatments for childhoodretinoblastoma.By McDaid C, Hartley S, BagnallA-M, Ritchie G, Light K, Riemsma R.No. 49Towards evidence-based guidelinesfor the prevention of venousthromboembolism: systematicreviews of mechanical methods, oralanticoagulation, dextran and regionalanaesthesia as thromboprophylaxis.By Roderick P, Ferris G, Wilson K,Halls H, Jackson D, Collins R, et al.No. 50The effectiveness and cost-effectivenessof parent training/educationprogrammes for the treatmentof conduct disorder, includingoppositional defiant disorder, inchildren.By Dretzke J, Frew E, Davenport C,Barlow J, Stewart-Brown S, Sandercock J,et al.Volume 10, 2006No. 1The clinical and cost-effectiveness ofdonepezil, rivastigmine, galantamineand memantine for Alzheimer’sdisease.By Loveman E, Green C, Kirby J,Takeda A, Picot J, Payne E, et al.No. 2FOOD: a multicentre randomised trialevaluating feeding policies in patientsadmitted to hospital with a recentstroke.By Dennis M, Lewis S, Cranswick G,Forbes J.No. 3The clinical effectiveness and costeffectivenessof computed tomographyscreening for lung cancer: systematicreviews.By Black C, Bagust A, Boland A,Walker S, McLeod C, De Verteuil R, et al.No. 4A systematic review of the effectivenessand cost-effectiveness of neuroimagingassessments used to visualise the seizurefocus in people with refractory epilepsybeing considered for surgery.By Whiting P, Gupta R, Burch J,Mujica Mota RE, Wright K, Marson A,et al.No. 5Comparison of conference abstractsand presentations with full-text articlesin the health technology assessments ofrapidly evolving technologies.By Dundar Y, Dodd S, Dickson R,Walley T, Haycox A, Williamson PR.No. 6Systematic review and evaluationof methods of assessing urinaryincontinence.By Martin JL, Williams KS, AbramsKR, Turner DA, Sutton AJ, Chapple C,et al.No. 7The clinical effectiveness and costeffectivenessof newer drugs forchildren with epilepsy. A systematicreview.By Connock M, Frew E, Evans B-W,Bryan S, Cummins C, Fry-Smith A, et al.No. 8Surveillance of Barrett’s oesophagus:exploring the uncertainty throughsystematic review, expert workshop andeconomic modelling.By Garside R, Pitt M, Somerville M,Stein K, Price A, Gilbert N.No. 9Topotecan, pegylated liposomaldoxorubicin hydrochloride andpaclitaxel for second-line or subsequenttreatment of advanced ovarian cancer:a systematic review and economicevaluation.By Main C, Bojke L, Griffin S,Norman G, Barbieri M, Mather L, et al.No. 10Evaluation of molecular techniquesin prediction and diagnosisof cytomegalovirus disease inimmunocompromised patients.By Szczepura A, Westmoreland D,Vinogradova Y, Fox J, Clark M.No. 11Screening for thrombophilia in highrisksituations: systematic reviewand cost-effectiveness analysis. TheThrombosis: Risk and EconomicAssessment of ThrombophiliaScreening (TREATS) study.By Wu O, Robertson L, Twaddle S,Lowe GDO, Clark P, Greaves M, et al.No. 12A series of systematic reviews to informa decision analysis for sampling andtreating infected diabetic foot ulcers.By Nelson EA, O’Meara S, Craig D,Iglesias C, Golder S, Dalton J, et al.No. 13Randomised clinical trial, observationalstudy and assessment of costeffectivenessof the treatment ofvaricose veins (REACTIV trial).By Michaels JA, Campbell WB,Brazier JE, MacIntyre JB, Palfreyman SJ,Ratcliffe J, et al.No. 14The cost-effectiveness of screening fororal cancer in primary care.By Speight PM, Palmer S, Moles DR,Downer MC, Smith DH, Henriksson M,et al.No. 15Measurement of the clinical and costeffectivenessof non-invasive diagnostictesting strategies for deep veinthrombosis.By Goodacre S, Sampson F,Stevenson M, Wailoo A, Sutton A,Thomas S, et al.No. 16Systematic review of the effectivenessand cost-effectiveness of HealOzone ®for the treatment of occlusal pit/fissurecaries and root caries.By Brazzelli M, McKenzie L, FieldingS, Fraser C, Clarkson J, Kilonzo M, et al.No. 17Randomised controlled trials ofconventional antipsychotic versusnew atypical drugs, and new atypicaldrugs versus clozapine, in people withschizophrenia responding poorly to, orintolerant of, current drug treatment.By Lewis SW, Davies L, Jones PB,Barnes TRE, Murray RM, Kerwin R,et al.No. 18Diagnostic tests and algorithms usedin the investigation of haematuria:systematic reviews and economicevaluation.By Rodgers M, Nixon J, Hempel S,Aho T, Kelly J, Neal D, et al.No. 19Cognitive behavioural therapy inaddition to antispasmodic therapy forirritable bowel syndrome in primarycare: randomised controlled trial.By Kennedy TM, Chalder T,McCrone P, Darnley S, Knapp M,Jones RH, et al.No. 20A systematic review of theclinical effectiveness and costeffectivenessof enzyme replacementtherapies for Fabry’s disease andmucopolysaccharidosis type 1.By Connock M, Juarez-Garcia A,Frew E, Mans A, Dretzke J, Fry-Smith A,et al.No. 21Health benefits of antiviral therapy formild chronic hepatitis C: randomisedcontrolled trial and economicevaluation.By Wright M, Grieve R, Roberts J,Main J, Thomas HC, on behalf of theUK Mild Hepatitis C Trial Investigators.No. 22Pressure relieving support surfaces: arandomised evaluation.By Nixon J, Nelson EA, Cranny G,Iglesias CP, Hawkins K, Cullum NA, et al.

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4No. 23A systematic review and economicmodel of the effectiveness and costeffectivenessof methylphenidate,dexamfetamine and atomoxetinefor the treatment of attention deficithyperactivity disorder in children andadolescents.By King S, Griffin S, Hodges Z,Weatherly H, Asseburg C, Richardson G,et al.No. 24The clinical effectiveness and costeffectivenessof enzyme replacementtherapy for Gaucher’s disease: asystematic review.By Connock M, Burls A, Frew E,Fry-Smith A, Juarez-Garcia A, McCabe C,et al.No. 25Effectiveness and cost-effectivenessof salicylic acid and cryotherapy forcutaneous warts. An economic decisionmodel.By Thomas KS, Keogh-Brown MR,Chalmers JR, Fordham RJ, Holland RC,Armstrong SJ, et al.No. 26A systematic literature review of theeffectiveness of non-pharmacologicalinterventions to prevent wandering indementia and evaluation of the ethicalimplications and acceptability of theiruse.By Robinson L, Hutchings D, CornerL, Beyer F, Dickinson H, Vanoli A, et al.No. 27A review of the evidence on the effectsand costs of implantable cardioverterdefibrillator therapy in differentpatient groups, and modelling of costeffectivenessand cost–utility for thesegroups in a UK context.By Buxton M, Caine N, Chase D,Connelly D, Grace A, Jackson C, et al.No. 28Adefovir dipivoxil and pegylatedinterferon alfa-2a for the treatment ofchronic hepatitis B: a systematic reviewand economic evaluation.By Shepherd J, Jones J, Takeda A,Davidson P, Price A.No. 29An evaluation of the clinical and costeffectivenessof pulmonary arterycatheters in patient management inintensive care: a systematic review and arandomised controlled trial.By Harvey S, Stevens K, Harrison D,Young D, Brampton W, McCabe C, et al.No. 30Accurate, practical and cost-effectiveassessment of carotid stenosis in theUK.By Wardlaw JM, Chappell FM,Stevenson M, De Nigris E, Thomas S,Gillard J, et al.No. 31Etanercept and infliximab for thetreatment of psoriatic arthritis: asystematic review and economicevaluation.By Woolacott N, Bravo Vergel Y,Hawkins N, Kainth A, Khadjesari Z,Misso K, et al.No. 32The cost-effectiveness of testing forhepatitis C in former injecting drugusers.By Castelnuovo E, Thompson-CoonJ, Pitt M, Cramp M, Siebert U, Price A,et al.No. 33Computerised cognitive behaviourtherapy for depression and anxietyupdate: a systematic review andeconomic evaluation.By Kaltenthaler E, Brazier J,De Nigris E, Tumur I, Ferriter M,Beverley C, et al.No. 34Cost-effectiveness of using prognosticinformation to select women with breastcancer for adjuvant systemic therapy.By Williams C, Brunskill S, Altman D,Briggs A, Campbell H, Clarke M, et al.No. 35Psychological therapies includingdialectical behaviour therapy forborderline personality disorder: asystematic review and preliminaryeconomic evaluation.By Brazier J, Tumur I, Holmes M,Ferriter M, Parry G, Dent-Brown K, et al.No. 36Clinical effectiveness and costeffectivenessof tests for the diagnosisand investigation of urinary tractinfection in children: a systematicreview and economic model.By Whiting P, Westwood M, Bojke L,Palmer S, Richardson G, Cooper J, et al.No. 37Cognitive behavioural therapyin chronic fatigue syndrome: arandomised controlled trial of anoutpatient group programme.By O’Dowd H, Gladwell P, RogersCA, Hollinghurst S, Gregory A.No. 38A comparison of the cost-effectivenessof five strategies for the preventionof nonsteroidal anti-inflammatorydrug-induced gastrointestinal toxicity:a systematic review with economicmodelling.By Brown TJ, Hooper L, Elliott RA,Payne K, Webb R, Roberts C, et al.No. 39The effectiveness and cost-effectivenessof computed tomography screeningfor coronary artery disease: systematicreview.By Waugh N, Black C, Walker S,McIntyre L, Cummins E, Hillis G.No. 40What are the clinical outcome and costeffectivenessof endoscopy undertakenby nurses when compared with doctors?A Multi-Institution Nurse EndoscopyTrial (MINuET).By Williams J, Russell I, Durai D,Cheung W-Y, Farrin A, Bloor K, et al.No. 41The clinical and cost-effectiveness ofoxaliplatin and capecitabine for theadjuvant treatment of colon cancer:systematic review and economicevaluation.By Pandor A, Eggington S, Paisley S,Tappenden P, Sutcliffe P.No. 42A systematic review of the effectivenessof adalimumab, etanercept andinfliximab for the treatment ofrheumatoid arthritis in adults andan economic evaluation of their costeffectiveness.By Chen Y-F, Jobanputra P, Barton P,Jowett S, Bryan S, Clark W, et al.No. 43Telemedicine in dermatology: arandomised controlled trial.By Bowns IR, Collins K, Walters SJ,McDonagh AJG.No. 44Cost-effectiveness of cell salvage andalternative methods of minimisingperioperative allogeneic bloodtransfusion: a systematic review andeconomic model.By Davies L, Brown TJ, Haynes S,Payne K, Elliott RA, McCollum C.No. 45Clinical effectiveness and costeffectivenessof laparoscopic surgeryfor colorectal cancer: systematic reviewsand economic evaluation.By Murray A, Lourenco T, de VerteuilR, Hernandez R, Fraser C, McKinley A,et al.No. 46Etanercept and efalizumab for thetreatment of psoriasis: a systematicreview.By Woolacott N, Hawkins N,Mason A, Kainth A, Khadjesari Z, BravoVergel Y, et al.No. 47Systematic reviews of clinical decisiontools for acute abdominal pain.By Liu JLY, Wyatt JC, Deeks JJ,Clamp S, Keen J, Verde P, et al.No. 48Evaluation of the ventricular assistdevice programme in the UK.By Sharples L, Buxton M, Caine N,Cafferty F, Demiris N, Dyer M, et al.345© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Health Technology Assessment reports published to date346No. 49A systematic review and economicmodel of the clinical and costeffectivenessof immunosuppressivetherapy for renal transplantation inchildren.By Yao G, Albon E, Adi Y, Milford D,Bayliss S, Ready A, et al.No. 50Amniocentesis results: investigation ofanxiety. The ARIA trial.By Hewison J, Nixon J, Fountain J,Cocks K, Jones C, Mason G, et al.Volume 11, 2007No. 1Pemetrexed disodium for the treatmentof malignant pleural mesothelioma:a systematic review and economicevaluation.By Dundar Y, Bagust A, Dickson R,Dodd S, Green J, Haycox A, et al.No. 2A systematic review and economicmodel of the clinical effectivenessand cost-effectiveness of docetaxelin combination with prednisone orprednisolone for the treatment ofhormone-refractory metastatic prostatecancer.By Collins R, Fenwick E, Trowman R,Perard R, Norman G, Light K, et al.No. 3A systematic review of rapid diagnostictests for the detection of tuberculosisinfection.By Dinnes J, Deeks J, Kunst H,Gibson A, Cummins E, Waugh N, et al.No. 4The clinical effectiveness and costeffectivenessof strontium ranelate forthe prevention of osteoporotic fragilityfractures in postmenopausal women.By Stevenson M, Davis S, Lloyd-JonesM, Beverley C.No. 5A systematic review of quantitative andqualitative research on the role andeffectiveness of written informationavailable to patients about individualmedicines.By Raynor DK, BlenkinsoppA, Knapp P, Grime J, Nicolson DJ,Pollock K, et al.No. 6Oral naltrexone as a treatment forrelapse prevention in formerly opioiddependentdrug users: a systematicreview and economic evaluation.By Adi Y, Juarez-Garcia A, Wang D,Jowett S, Frew E, Day E, et al.No. 7Glucocorticoid-induced osteoporosis:a systematic review and cost–utilityanalysis.By Kanis JA, Stevenson M,McCloskey EV, Davis S, Lloyd-Jones M.No. 8Epidemiological, social, diagnostic andeconomic evaluation of populationscreening for genital chlamydialinfection.By Low N, McCarthy A, Macleod J,Salisbury C, Campbell R, Roberts TE,et al.No. 9Methadone and buprenorphine for themanagement of opioid dependence:a systematic review and economicevaluation.By Connock M, Juarez-Garcia A,Jowett S, Frew E, Liu Z, Taylor RJ, et al.No. 10Exercise Evaluation RandomisedTrial (EXERT): a randomised trialcomparing GP referral for leisurecentre-based exercise, community-basedwalking and advice only.By Isaacs AJ, Critchley JA, See TaiS, Buckingham K, Westley D, HarridgeSDR, et al.No. 11Interferon alfa (pegylated and nonpegylated)and ribavirin for thetreatment of mild chronic hepatitisC: a systematic review and economicevaluation.By Shepherd J, Jones J, Hartwell D,Davidson P, Price A, Waugh N.No. 12Systematic review and economicevaluation of bevacizumab andcetuximab for the treatment ofmetastatic colorectal cancer.By Tappenden P, Jones R, Paisley S,Carroll C.No. 13A systematic review and economicevaluation of epoetin alfa, epoetinbeta and darbepoetin alfa in anaemiaassociated with cancer, especially thatattributable to cancer treatment.By Wilson J, Yao GL, Raftery J,Bohlius J, Brunskill S, Sandercock J,et al.No. 14A systematic review and economicevaluation of statins for the preventionof coronary events.By Ward S, Lloyd Jones M, Pandor A,Holmes M, Ara R, Ryan A, et al.No. 15A systematic review of the effectivenessand cost-effectiveness of differentmodels of community-based respitecare for frail older people and theircarers.By Mason A, Weatherly H, SpilsburyK, Arksey H, Golder S, Adamson J, et al.No. 16Additional therapy for youngchildren with spastic cerebral palsy: arandomised controlled trial.By Weindling AM, Cunningham CC,Glenn SM, Edwards RT, Reeves DJ.No. 17Screening for type 2 diabetes: literaturereview and economic modelling.By Waugh N, Scotland G, McNameeP, Gillett M, Brennan A, Goyder E, et al.No. 18The effectiveness and cost-effectivenessof cinacalcet for secondaryhyperparathyroidism in end-stage renaldisease patients on dialysis: a systematicreview and economic evaluation.By Garside R, Pitt M, Anderson R,Mealing S, Roome C, Snaith A, et al.No. 19The clinical effectiveness and costeffectivenessof gemcitabine formetastatic breast cancer: a systematicreview and economic evaluation.By Takeda AL, Jones J, Loveman E,Tan SC, Clegg AJ.No. 20A systematic review of duplexultrasound, magnetic resonanceangiography and computedtomography angiography forthe diagnosis and assessment ofsymptomatic, lower limb peripheralarterial disease.By Collins R, Cranny G, Burch J,Aguiar-Ibáñez R, Craig D, Wright K,et al.No. 21The clinical effectiveness and costeffectivenessof treatments for childrenwith idiopathic steroid-resistantnephrotic syndrome: a systematicreview.By Colquitt JL, Kirby J, Green C,Cooper K, Trompeter RS.No. 22A systematic review of the routinemonitoring of growth in children ofprimary school age to identify growthrelatedconditions.By Fayter D, Nixon J, Hartley S,Rithalia A, Butler G, Rudolf M, et al.No. 23Systematic review of the effectiveness ofpreventing and treating Staphylococcusaureus carriage in reducing peritonealcatheter-related infections.By McCormack K, Rabindranath K,Kilonzo M, Vale L, Fraser C, McIntyre L,et al.

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4No. 24The clinical effectiveness and costof repetitive transcranial magneticstimulation versus electroconvulsivetherapy in severe depression: amulticentre pragmatic randomisedcontrolled trial and economic analysis.By McLoughlin DM, Mogg A, ErantiS, Pluck G, Purvis R, Edwards D, et al.No. 25A randomised controlled trial andeconomic evaluation of direct versusindirect and individual versus groupmodes of speech and language therapyfor children with primary languageimpairment.By Boyle J, McCartney E, Forbes J,O’Hare A.No. 26Hormonal therapies for early breastcancer: systematic review and economicevaluation.By Hind D, Ward S, De Nigris E,Simpson E, Carroll C, Wyld L.No. 27Cardioprotection against the toxiceffects of anthracyclines given tochildren with cancer: a systematicreview.By Bryant J, Picot J, Levitt G,Sullivan I, Baxter L, Clegg A.No. 28Adalimumab, etanercept and infliximabfor the treatment of ankylosingspondylitis: a systematic review andeconomic evaluation.By McLeod C, Bagust A, Boland A,Dagenais P, Dickson R, Dundar Y, et al.No. 29Prenatal screening and treatmentstrategies to prevent group Bstreptococcal and other bacterialinfections in early infancy: costeffectivenessand expected value ofinformation analyses.By Colbourn T, Asseburg C, Bojke L,Philips Z, Claxton K, Ades AE, et al.No. 30Clinical effectiveness and costeffectivenessof bone morphogeneticproteins in the non-healing of fracturesand spinal fusion: a systematic review.By Garrison KR, Donell S, Ryder J,Shemilt I, Mugford M, Harvey I, et al.No. 31A randomised controlled trial ofpostoperative radiotherapy followingbreast-conserving surgery in aminimum-risk older population. ThePRIME trial.By Prescott RJ, Kunkler IH, WilliamsLJ, King CC, Jack W, van der Pol M,et al.No. 32Current practice, accuracy, effectivenessand cost-effectiveness of the schoolentry hearing screen.By Bamford J, Fortnum H, BristowK, Smith J, Vamvakas G, Davies L, et al.No. 33The clinical effectiveness and costeffectivenessof inhaled insulin indiabetes mellitus: a systematic reviewand economic evaluation.By Black C, Cummins E, Royle P,Philip S, Waugh N.No. 34Surveillance of cirrhosis forhepatocellular carcinoma: systematicreview and economic analysis.By Thompson Coon J, Rogers G,Hewson P, Wright D, Anderson R,Cramp M, et al.No. 35The Birmingham RehabilitationUptake Maximisation Study (BRUM).Homebased compared with hospitalbasedcardiac rehabilitation in a multiethnicpopulation: cost-effectivenessand patient adherence.By Jolly K, Taylor R, Lip GYH,Greenfield S, Raftery J, Mant J, et al.No. 36A systematic review of the clinical,public health and cost-effectiveness ofrapid diagnostic tests for the detectionand identification of bacterial intestinalpathogens in faeces and food.By Abubakar I, Irvine L, Aldus CF,Wyatt GM, Fordham R, Schelenz S, et al.No. 37A randomised controlled trialexamining the longer-term outcomesof standard versus new antiepilepticdrugs. The SANAD trial.By Marson AG, Appleton R, BakerGA, Chadwick DW, Doughty J, Eaton B,et al.No. 38Clinical effectiveness and costeffectivenessof different modelsof managing long-term oral anticoagulationtherapy: a systematicreview and economic modelling.By Connock M, Stevens C, Fry-SmithA, Jowett S, Fitzmaurice D, Moore D,et al.No. 39A systematic review and economicmodel of the clinical effectivenessand cost-effectiveness of interventionsfor preventing relapse in people withbipolar disorder.By Soares-Weiser K, Bravo Vergel Y,Beynon S, Dunn G, Barbieri M, Duffy S,et al.No. 40Taxanes for the adjuvant treatment ofearly breast cancer: systematic reviewand economic evaluation.By Ward S, Simpson E, Davis S, HindD, Rees A, Wilkinson A.No. 41The clinical effectiveness and costeffectivenessof screening for openangle glaucoma: a systematic reviewand economic evaluation.By Burr JM, Mowatt G, HernándezR, Siddiqui MAR, Cook J, Lourenco T,et al.No. 42Acceptability, benefit and costs of earlyscreening for hearing disability: a studyof potential screening tests and models.By Davis A, Smith P, Ferguson M,Stephens D, Gianopoulos I.No. 43Contamination in trials of educationalinterventions.By Keogh-Brown MR, BachmannMO, Shepstone L, Hewitt C, Howe A,Ramsay CR, et al.No. 44Overview of the clinical effectiveness ofpositron emission tomography imagingin selected cancers.By Facey K, Bradbury I, Laking G,Payne E.No. 45The effectiveness and cost-effectivenessof carmustine implants andtemozolomide for the treatment ofnewly diagnosed high-grade glioma:a systematic review and economicevaluation.By Garside R, Pitt M, Anderson R,Rogers G, Dyer M, Mealing S, et al.No. 46Drug-eluting stents: a systematic reviewand economic evaluation.By Hill RA, Boland A, Dickson R,Dündar Y, Haycox A, McLeod C, et al.No. 47The clinical effectiveness andcost-effectiveness of cardiacresynchronisation (biventricular pacing)for heart failure: systematic review andeconomic model.By Fox M, Mealing S, Anderson R,Dean J, Stein K, Price A, et al.No. 48Recruitment to randomised trials:strategies for trial enrolment andparticipation study. The STEPS study.By Campbell MK, Snowdon C,Francis D, Elbourne D, McDonald AM,Knight R, et al.347© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Health Technology Assessment reports published to dateNo. 49Cost-effectiveness of functionalcardiac testing in the diagnosis andmanagement of coronary arterydisease: a randomised controlled trial.The CECaT trial.By Sharples L, Hughes V, Crean A,Dyer M, Buxton M, Goldsmith K, et al.No. 50Evaluation of diagnostic tests whenthere is no gold standard. A review ofmethods.By Rutjes AWS, ReitsmaJB, Coomarasamy A, Khan KS,Bossuyt PMM.No. 51Systematic reviews of the clinicaleffectiveness and cost-effectiveness ofproton pump inhibitors in acute uppergastrointestinal bleeding.By Leontiadis GI, SreedharanA, Dorward S, Barton P, Delaney B,Howden CW, et al.No. 52A review and critique of modelling inprioritising and designing screeningprogrammes.By Karnon J, Goyder E, TappendenP, McPhie S, Towers I, Brazier J, et al.No. 53An assessment of the impact of theNHS Health Technology AssessmentProgramme.By Hanney S, Buxton M, Green C,Coulson D, Raftery J.Volume 12, 2008No. 1A systematic review and economicmodel of switching fromnonglycopeptide to glycopeptideantibiotic prophylaxis for surgery.By Cranny G, Elliott R, Weatherly H,Chambers D, Hawkins N, Myers L, et al.No. 4Does befriending by trained lay workersimprove psychological well-being andquality of life for carers of peoplewith dementia, and at what cost? Arandomised controlled trial.By Charlesworth G, Shepstone L,Wilson E, Thalanany M, Mugford M,Poland F.No. 5A multi-centre retrospective cohortstudy comparing the efficacy, safetyand cost-effectiveness of hysterectomyand uterine artery embolisation forthe treatment of symptomatic uterinefibroids. The HOPEFUL study.By Hirst A, Dutton S, Wu O, BriggsA, Edwards C, Waldenmaier L, et al.No. 6Methods of prediction and preventionof pre-eclampsia: systematic reviews ofaccuracy and effectiveness literaturewith economic modelling.By Meads CA, Cnossen JS, Meher S,Juarez-Garcia A, ter Riet G, Duley L,et al.No. 7The use of economic evaluations inNHS decision-making: a review andempirical investigation.By Williams I, McIver S, Moore D,Bryan S.No. 8Stapled haemorrhoidectomy(haemorrhoidopexy) for the treatmentof haemorrhoids: a systematic reviewand economic evaluation.By Burch J, Epstein D, Baba-AkbariA, Weatherly H, Fox D, Golder S, et al.No. 9The clinical effectiveness of diabeteseducation models for Type 2 diabetes: asystematic review.By Loveman E, Frampton GK,Clegg AJ.No. 12The clinical effectiveness and costeffectivenessof central venous catheterstreated with anti-infective agents inpreventing bloodstream infections:a systematic review and economicevaluation.By Hockenhull JC, Dwan K, BolandA, Smith G, Bagust A, Dundar Y, et al.No. 13Stepped treatment of older adults onlaxatives. The STOOL trial.By Mihaylov S, Stark C, McColl E,Steen N, Vanoli A, Rubin G, et al.No. 14A randomised controlled trial ofcognitive behaviour therapy inadolescents with major depressiontreated by selective serotonin reuptakeinhibitors. The ADAPT trial.By Goodyer IM, Dubicka B,Wilkinson P, Kelvin R, Roberts C,Byford S, et al.No. 15The use of irinotecan, oxaliplatinand raltitrexed for the treatment ofadvanced colorectal cancer: systematicreview and economic evaluation.By Hind D, Tappenden P, Tumur I,Eggington E, Sutcliffe P, Ryan A.No. 16Ranibizumab and pegaptanib forthe treatment of age-related maculardegeneration: a systematic review andeconomic evaluation.By Colquitt JL, Jones J, Tan SC,Takeda A, Clegg AJ, Price A.No. 17Systematic review of the clinicaleffectiveness and cost-effectivenessof 64-slice or higher computedtomography angiography as analternative to invasive coronaryangiography in the investigation ofcoronary artery disease.By Mowatt G, Cummins E, Waugh N,Walker S, Cook J, Jia X, et al.348No. 2‘Cut down to quit’ with nicotinereplacement therapies in smokingcessation: a systematic review ofeffectiveness and economic analysis.By Wang D, Connock M, Barton P,Fry-Smith A, Aveyard P, Moore D.No. 3A systematic review of the effectivenessof strategies for reducing fracture riskin children with juvenile idiopathicarthritis with additional data on longtermrisk of fracture and cost of diseasemanagement.By Thornton J, Ashcroft D, O’Neill T,Elliott R, Adams J, Roberts C, et al.No. 10Payment to healthcare professionals forpatient recruitment to trials: systematicreview and qualitative study.By Raftery J, Bryant J, Powell J,Kerr C, Hawker S.No. 11Cyclooxygenase-2 selective nonsteroidalanti-inflammatory drugs(etodolac, meloxicam, celecoxib,rofecoxib, etoricoxib, valdecoxib andlumiracoxib) for osteoarthritis andrheumatoid arthritis: a systematicreview and economic evaluation.By Chen Y-F, Jobanputra P, Barton P,Bryan S, Fry-Smith A, Harris G, et al.No. 18Structural neuroimaging in psychosis:a systematic review and economicevaluation.By Albon E, Tsourapas A, Frew E,Davenport C, Oyebode F, Bayliss S, et al.No. 19Systematic review and economicanalysis of the comparativeeffectiveness of different inhaledcorticosteroids and their usage withlong-acting beta 2agonists for thetreatment of chronic asthma in adultsand children aged 12 years and over.By Shepherd J, Rogers G, AndersonR, Main C, Thompson-Coon J,Hartwell D, et al.

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4No. 20Systematic review and economicanalysis of the comparativeeffectiveness of different inhaledcorticosteroids and their usage withlong-acting beta 2agonists for thetreatment of chronic asthma in childrenunder the age of 12 years.By Main C, Shepherd J, Anderson R,Rogers G, Thompson-Coon J, Liu Z,et al.No. 21Ezetimibe for the treatment ofhypercholesterolaemia: a systematicreview and economic evaluation.By Ara R, Tumur I, Pandor A,Duenas A, Williams R, Wilkinson A, et al.No. 22Topical or oral ibuprofen for chronicknee pain in older people. The TOIBstudy.By Underwood M, Ashby D, CarnesD, Castelnuovo E, Cross P, Harding G,et al.No. 23A prospective randomised comparisonof minor surgery in primary andsecondary care. The MiSTIC trial.By George S, Pockney P, Primrose J,Smith H, Little P, Kinley H, et al.No. 24A review and critical appraisalof measures of therapist–patientinteractions in mental health settings.By Cahill J, Barkham M, Hardy G,Gilbody S, Richards D, Bower P, et al.No. 25The clinical effectiveness and costeffectivenessof screening programmesfor amblyopia and strabismus inchildren up to the age of 4–5 years:a systematic review and economicevaluation.By Carlton J, Karnon J, Czoski-Murray C, Smith KJ, Marr J.No. 26A systematic review of the clinicaleffectiveness and cost-effectivenessand economic modelling of minimalincision total hip replacementapproaches in the management ofarthritic disease of the hip.By de Verteuil R, Imamura M, Zhu S,Glazener C, Fraser C, Munro N, et al.No. 27A preliminary model-based assessmentof the cost–utility of a screeningprogramme for early age-relatedmacular degeneration.By Karnon J, Czoski-Murray C,Smith K, Brand C, Chakravarthy U,Davis S, et al.No. 28Intravenous magnesium sulphateand sotalol for prevention of atrialfibrillation after coronary arterybypass surgery: a systematic review andeconomic evaluation.By Shepherd J, Jones J, FramptonGK, Tanajewski L, Turner D, Price A.No. 29Absorbent products for urinary/faecalincontinence: a comparative evaluationof key product categories.By Fader M, Cottenden A, Getliffe K,Gage H, Clarke-O’Neill S, Jamieson K,et al.No. 30A systematic review of repetitivefunctional task practice with modellingof resource use, costs and effectiveness.By French B, Leathley M, Sutton C,McAdam J, Thomas L, Forster A, et al.No. 31The effectiveness and cost-effectivnessof minimal access surgery amongstpeople with gastro-oesophageal refluxdisease – a UK collaborative study. Thereflux trial.By Grant A, Wileman S, Ramsay C,Bojke L, Epstein D, Sculpher M, et al.No. 32Time to full publication of studies ofanti-cancer medicines for breast cancerand the potential for publication bias: ashort systematic review.By Takeda A, Loveman E, Harris P,Hartwell D, Welch K.No. 33Performance of screening tests forchild physical abuse in accident andemergency departments.By Woodman J, Pitt M, Wentz R,Taylor B, Hodes D, Gilbert RE.No. 34Curative catheter ablation in atrialfibrillation and typical atrial flutter:systematic review and economicevaluation.By Rodgers M, McKenna C, PalmerS, Chambers D, Van Hout S, Golder S,et al.No. 35Systematic review and economicmodelling of effectiveness and costutility of surgical treatments for menwith benign prostatic enlargement.By Lourenco T, Armstrong N, N’DowJ, Nabi G, Deverill M, Pickard R, et al.No. 36Immunoprophylaxis against respiratorysyncytial virus (RSV) with palivizumabin children: a systematic review andeconomic evaluation.By Wang D, Cummins C, Bayliss S,Sandercock J, Burls A.Volume 13, 2009No. 1Deferasirox for the treatment of ironoverload associated with regularblood transfusions (transfusionalhaemosiderosis) in patients sufferingwith chronic anaemia: a systematicreview and economic evaluation.By McLeod C, Fleeman N, KirkhamJ, Bagust A, Boland A, Chu P, et al.No. 2Thrombophilia testing in people withvenous thromboembolism: systematicreview and cost-effectiveness analysis.By Simpson EL, Stevenson MD,Rawdin A, Papaioannou D.No. 3Surgical procedures and non-surgicaldevices for the management of nonapnoeicsnoring: a systematic review ofclinical effects and associated treatmentcosts.By Main C, Liu Z, Welch K, WeinerG, Quentin Jones S, Stein K.No. 4Continuous positive airway pressuredevices for the treatment of obstructivesleep apnoea–hypopnoea syndrome: asystematic review and economic analysis.By McDaid C, Griffin S, Weatherly H,Durée K, van der Burgt M, van Hout S,Akers J, et al.No. 5Use of classical and novel biomarkersas prognostic risk factors for localisedprostate cancer: a systematic review.By Sutcliffe P, Hummel S, Simpson E,Young T, Rees A, Wilkinson A, et al.No. 6The harmful health effects ofrecreational ecstasy: a systematic reviewof observational evidence.By Rogers G, Elston J, Garside R,Roome C, Taylor R, Younger P, et al.No. 7Systematic review of the clinicaleffectiveness and cost-effectivenessof oesophageal Doppler monitoringin critically ill and high-risk surgicalpatients.By Mowatt G, Houston G, HernándezR, de Verteuil R, Fraser C, CuthbertsonB, et al.No. 8The use of surrogate outcomes inmodel-based cost-effectiveness analyses:a survey of UK Health TechnologyAssessment reports.By Taylor RS, Elston J.No. 9Controlling Hypertension andHypotension Immediately Post Stroke(CHHIPS) – a randomised controlledtrial.By Potter J, Mistri A, Brodie F,Chernova J, Wilson E, Jagger C, et al.349© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Health Technology Assessment reports published to date350No. 10Routine antenatal anti-D prophylaxisfor RhD-negative women: a systematicreview and economic evaluation.By Pilgrim H, Lloyd-Jones M, Rees A.No. 11Amantadine, oseltamivir and zanamivirfor the prophylaxis of influenza(including a review of existing guidanceno. 67): a systematic review andeconomic evaluation.By Tappenden P, Jackson R, CooperK, Rees A, Simpson E, Read R, et al.No. 12Improving the evaluation oftherapeutic interventions in multiplesclerosis: the role of new psychometricmethods.By Hobart J, Cano S.No. 13Treatment of severe ankle sprain: apragmatic randomised controlled trialcomparing the clinical effectivenessand cost-effectiveness of three types ofmechanical ankle support with tubularbandage. The CAST trial.By Cooke MW, Marsh JL, Clark M,Nakash R, Jarvis RM, Hutton JL, et al.,on behalf of the CAST trial group.No. 14Non-occupational postexposureprophylaxis for HIV: a systematicreview.By Bryant J, Baxter L, Hird S.No. 15Blood glucose self-monitoring in type 2diabetes: a randomised controlled trial.By Farmer AJ, Wade AN, French DP,Simon J, Yudkin P, Gray A, et al.No. 16How far does screening women fordomestic (partner) violence in differenthealth-care settings meet criteria fora screening programme? Systematicreviews of nine UK National ScreeningCommittee criteria.By Feder G, Ramsay J, Dunne D,Rose M, Arsene C, Norman R, et al.No. 17Spinal cord stimulation for chronicpain of neuropathic or ischaemicorigin: systematic review and economicevaluation.By Simpson, EL, Duenas A, HolmesMW, Papaioannou D, Chilcott J.No. 18The role of magnetic resonanceimaging in the identification ofsuspected acoustic neuroma: asystematic review of clinical and costeffectivenessand natural history.By Fortnum H, O’Neill C, Taylor R,Lenthall R, Nikolopoulos T, LightfootG, et al.No. 19Dipsticks and diagnostic algorithms inurinary tract infection: developmentand validation, randomised trial,economic analysis, observational cohortand qualitative study.By Little P, Turner S, Rumsby K,Warner G, Moore M, Lowes JA, et al.No. 20Systematic review of respite care in thefrail elderly.By Shaw C, McNamara R, AbramsK, Cannings-John R, Hood K, LongoM, et al.No. 21Neuroleptics in the treatment ofaggressive challenging behaviour forpeople with intellectual disabilities:a randomised controlled trial(NACHBID).By Tyrer P, Oliver-Africano P, RomeoR, Knapp M, Dickens S, Bouras N, et al.No. 22Randomised controlled trial todetermine the clinical effectivenessand cost-effectiveness of selectiveserotonin reuptake inhibitors plussupportive care, versus supportive carealone, for mild to moderate depressionwith somatic symptoms in primarycare: the THREAD (THREshold forAntiDepressant response) study.By Kendrick T, Chatwin J, Dowrick C,Tylee A, Morriss R, Peveler R, et al.No. 23Diagnostic strategies using DNA testingfor hereditary haemochromatosis inat-risk populations: a systematic reviewand economic evaluation.By Bryant J, Cooper K, Picot J, CleggA, Roderick P, Rosenberg W, et al.No. 24Enhanced external counterpulsationfor the treatment of stable angina andheart failure: a systematic review andeconomic analysis.By McKenna C, McDaid C,Suekarran S, Hawkins N, Claxton K,Light K, et al.No. 25Development of a decision supporttool for primary care management ofpatients with abnormal liver functiontests without clinically apparent liverdisease: a record-linkage populationcohort study and decision analysis(ALFIE).By Donnan PT, McLernon D, DillonJF, Ryder S, Roderick P, Sullivan F, et al.No. 26A systematic review of presumedconsent systems for deceased organdonation.By Rithalia A, McDaid C, SuekarranS, Norman G, Myers L, Sowden A.No. 27Paracetamol and ibuprofen for thetreatment of fever in children: thePITCH randomised controlled trial.By Hay AD, Redmond NM, CostelloeC, Montgomery AA, Fletcher M,Hollinghurst S, et al.No. 28A randomised controlled trial tocompare minimally invasive glucosemonitoring devices with conventionalmonitoring in the management ofinsulin-treated diabetes mellitus(MITRE).By Newman SP, Cooke D, Casbard A,Walker S, Meredith S, Nunn A, et al.No. 29Sensitivity analysis in economicevaluation: an audit of NICE currentpractice and a review of its use andvalue in decision-making.By Andronis L, Barton P, Bryan S.Suppl. 1Trastuzumab for the treatment ofprimary breast cancer in HER2-positivewomen: a single technology appraisal.By Ward S, Pilgrim H, Hind D.Docetaxel for the adjuvant treatmentof early node-positive breast cancer: asingle technology appraisal.By Chilcott J, Lloyd Jones M,Wilkinson A.The use of paclitaxel in themanagement of early stage breastcancer.By Griffin S, Dunn G, Palmer S,Macfarlane K, Brent S, Dyker A, et al.Rituximab for the first-line treatmentof stage III/IV follicular non-Hodgkin’slymphoma.By Dundar Y, Bagust A, Hounsome J,McLeod C, Boland A, Davis H, et al.Bortezomib for the treatment ofmultiple myeloma patients.By Green C, Bryant J, Takeda A,Cooper K, Clegg A, Smith A, et al.Fludarabine phosphate for the firstlinetreatment of chronic lymphocyticleukaemia.By Walker S, Palmer S, Erhorn S,Brent S, Dyker A, Ferrie L, et al.Erlotinib for the treatment of relapsednon-small cell lung cancer.By McLeod C, Bagust A, Boland A,Hockenhull J, Dundar Y, Proudlove C,et al.Cetuximab plus radiotherapy for thetreatment of locally advanced squamouscell carcinoma of the head and neck.By Griffin S, Walker S, Sculpher M,White S, Erhorn S, Brent S, et al.Infliximab for the treatment of adultswith psoriasis.By Loveman E, Turner D, HartwellD, Cooper K, Clegg A.

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4No. 30Psychological interventions forpostnatal depression: clusterrandomised trial and economicevaluation. The PoNDER trial.By Morrell CJ, Warner R, Slade P,Dixon S, Walters S, Paley G, et al.No. 31The effect of different treatmentdurations of clopidogrel in patientswith non-ST-segment elevation acutecoronary syndromes: a systematicreview and value of informationanalysis.By Rogowski R, Burch J, Palmer S,Craigs C, Golder S, Woolacott N.No. 32Systematic review and individualpatient data meta-analysis of diagnosisof heart failure, with modelling ofimplications of different diagnosticstrategies in primary care.By Mant J, Doust J, Roalfe A, BartonP, Cowie MR, Glasziou P, et al.No. 33A multicentre randomised controlledtrial of the use of continuous positiveairway pressure and non-invasivepositive pressure ventilation in the earlytreatment of patients presenting to theemergency department with severeacute cardiogenic pulmonary oedema:the 3CPO trial.By Gray AJ, Goodacre S, NewbyDE, Masson MA, Sampson F, DixonS, et al., on behalf of the 3CPO studyinvestigators.No. 34Early high-dose lipid-lowering therapyto avoid cardiac events: a systematicreview and economic evaluation.By Ara R, Pandor A, Stevens J, ReesA, Rafia R.No. 35Adefovir dipivoxil and pegylatedinterferon alpha for the treatmentof chronic hepatitis B: an updatedsystematic review and economicevaluation.By Jones J, Shepherd J, Baxter L,Gospodarevskaya E, Hartwell D, HarrisP, et al.No. 36Methods to identify postnataldepression in primary care: anintegrated evidence synthesis and valueof information analysis.By Hewitt CE, Gilbody SM, BrealeyS, Paulden M, Palmer S, Mann R, et al.No. 37A double-blind randomised placebocontrolledtrial of topical intranasalcorticosteroids in 4- to 11-year-oldchildren with persistent bilateral otitismedia with effusion in primary care.By Williamson I, Benge S, Barton S,Petrou S, Letley L, Fasey N, et al.No. 38The effectiveness and cost-effectivenessof methods of storing donated kidneysfrom deceased donors: a systematicreview and economic model.By Bond M, Pitt M, Akoh J, MoxhamT, Hoyle M, Anderson R.No. 39Rehabilitation of older patients: dayhospital compared with rehabilitationat home. A randomised controlled trial.By Parker SG, Oliver P, PenningtonM, Bond J, Jagger C, Enderby PM, et al.No. 40Breastfeeding promotion for infants inneonatal units: a systematic review andeconomic analysisBy Renfrew MJ, Craig D, Dyson L,McCormick F, Rice S, King SE, et al.No. 41The clinical effectiveness and costeffectivenessof bariatric (weight loss)surgery for obesity: a systematic review andeconomic evaluation.By Picot J, Jones J, Colquitt JL,Gospodarevskaya E, Loveman E, BaxterL, et al.No. 42Rapid testing for group B streptococcusduring labour: a test accuracy studywith evaluation of acceptability andcost-effectiveness.By Daniels J, Gray J, Pattison H,Roberts T, Edwards E, Milner P, et al.No. 43Screening to prevent spontaneouspreterm birth: systematic reviews ofaccuracy and effectiveness literaturewith economic modelling.By Honest H, Forbes CA, Durée KH,Norman G, Duffy SB, Tsourapas A, et al.No. 44The effectiveness and cost-effectivenessof cochlear implants for severe toprofound deafness in children andadults: a systematic review andeconomic model.By Bond M, Mealing S, Anderson R,Elston J, Weiner G, Taylor RS, et al.Suppl. 2Gemcitabine for the treatment ofmetastatic breast cancer.By Jones J, Takeda A, Tan SC,Cooper K, Loveman E, Clegg A.Varenicline in the management ofsmoking cessation: a single technologyappraisal.By Hind D, Tappenden P, Peters J,Kenjegalieva K.Alteplase for the treatment of acuteischaemic stroke: a single technologyappraisal.By Lloyd Jones M, Holmes M.Rituximab for the treatment ofrheumatoid arthritis.By Bagust A, Boland A, HockenhullJ, Fleeman N, Greenhalgh J, Dundar Y,et al.Omalizumab for the treatment ofsevere persistent allergic asthma.By Jones J, Shepherd J, Hartwell D,Harris P, Cooper K, Takeda A, et al.Rituximab for the treatment of relapsedor refractory stage III or IV follicularnon-Hodgkin’s lymphoma.By Boland A, Bagust A, HockenhullJ, Davis H, Chu P, Dickson R.Adalimumab for the treatment ofpsoriasis.By Turner D, Picot J, Cooper K,Loveman E.Dabigatran etexilate for the preventionof venous thromboembolism in patientsundergoing elective hip and kneesurgery: a single technology appraisal.By Holmes M, C Carroll C,Papaioannou D.Romiplostim for the treatmentof chronic immune or idiopathicthrombocytopenic purpura: a singletechnology appraisal.By Mowatt G, Boachie C, CrowtherM, Fraser C, Hernández R, Jia X, et al.Sunitinib for the treatment ofgastrointestinal stromal tumours: acritique of the submission from Pfizer.By Bond M, Hoyle M, Moxham T,Napier M, Anderson R.No. 45Vitamin K to prevent fractures inolder women: systematic review andeconomic evaluation.By Stevenson M, Lloyd-Jones M,Papaioannou D.No. 46The effects of biofeedback for thetreatment of essential hypertension: asystematic review.By Greenhalgh J, Dickson R,Dundar Y.No. 47A randomised controlled trial of theuse of aciclovir and/or prednisolone forthe early treatment of Bell’s palsy: theBELLS study.By Sullivan FM, Swan IRC, DonnanPT, Morrison JM, Smith BH, McKinstryB, et al.Suppl. 3Lapatinib for the treatment of HER2-overexpressing breast cancer.By Jones J, Takeda A, Picot J, vonKeyserlingk C, Clegg A.Infliximab for the treatment ofulcerative colitis.By Hyde C, Bryan S, Juarez-Garcia A,Andronis L, Fry-Smith A.351© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Health Technology Assessment reports published to date352Rimonabant for the treatment ofoverweight and obese people.By Burch J, McKenna C, Palmer S,Norman G, Glanville J, Sculpher M, etal.Telbivudine for the treatment ofchronic hepatitis B infection.By Hartwell D, Jones J, Harris P,Cooper K.Entecavir for the treatment of chronichepatitis B infection.By Shepherd J, Gospodarevskaya E,Frampton G, Cooper, K.Febuxostat for the treatment ofhyperuricaemia in people with gout: asingle technology appraisal.By Stevenson M, Pandor A.Rivaroxaban for the prevention ofvenous thromboembolism: a singletechnology appraisal.By Stevenson M, Scope A, Holmes M,Rees A, Kaltenthaler E.Cetuximab for the treatment ofrecurrent and/or metastatic squamouscell carcinoma of the head and neck.By Greenhalgh J, Bagust A, BolandA, Fleeman N, McLeod C, Dundar Y,et al.Mifamurtide for the treatment ofosteosarcoma: a single technologyappraisal.By Pandor A, Fitzgerald P, StevensonM, Papaioannou D.Ustekinumab for the treatment ofmoderate to severe psoriasis.By Gospodarevskaya E, Picot J,Cooper K, Loveman E, Takeda A.No. 48Endovascular stents for abdominalaortic aneurysms: a systematic reviewand economic model.By Chambers D, Epstein D, Walker S,Fayter D, Paton F, Wright K, et al.No. 49Clinical and cost-effectiveness ofepoprostenol, iloprost, bosentan,sitaxentan and sildenafil for pulmonaryarterial hypertension within theirlicensed indications: a systematic reviewand economic evaluation.By Chen Y-F, Jowett S, Barton P,Malottki K, Hyde C, Gibbs JSR, et al.No. 50Cessation of attention deficithyperactivity disorder drugsin the young (CADDY) – apharmacoepidemiological andqualitative study.By Wong ICK, Asherson P, Bilbow A,Clifford S, Coghill D, R DeSoysa R, et al.No. 51ARTISTIC: a randomised trial ofhuman papillomavirus (HPV) testing inprimary cervical screening.By Kitchener HC, Almonte M,Gilham C, Dowie R, Stoykova B, SargentA, et al.No. 52The clinical effectiveness ofglucosamine and chondroitinsupplements in slowing or arrestingprogression of osteoarthritis of theknee: a systematic review and economicevaluation.By Black C, Clar C, Henderson R,MacEachern C, McNamee P, QuayyumZ, et al.No. 53Randomised preference trial ofmedical versus surgical termination ofpregnancy less than 14 weeks’ gestation(TOPS).By Robson SC, Kelly T, Howel D,Deverill M, Hewison J, Lie MLS, et al.No. 54Randomised controlled trial of the useof three dressing preparations in themanagement of chronic ulceration ofthe foot in diabetes.By Jeffcoate WJ, Price PE, PhillipsCJ, Game FL, Mudge E, Davies S, et al.No. 55VenUS II: a randomised controlled trialof larval therapy in the management ofleg ulcers.By Dumville JC, Worthy G, SoaresMO, Bland JM, Cullum N, Dowson C,et al.No. 56A prospective randomised controlledtrial and economic modelling ofantimicrobial silver dressings versusnon-adherent control dressings forvenous leg ulcers: the VULCAN trialBy Michaels JA, Campbell WB,King BM, MacIntyre J, Palfreyman SJ,Shackley P, et al.No. 57Communication of carrier statusinformation following universalnewborn screening for sickle celldisorders and cystic fibrosis: qualitativestudy of experience and practice.By Kai J, Ulph F, Cullinan T,Qureshi N.No. 58Antiviral drugs for the treatment ofinfluenza: a systematic review andeconomic evaluation.By Burch J, Paulden M, Conti S,Stock C, Corbett M, Welton NJ, et al.No. 59Development of a toolkit and glossaryto aid in the adaptation of healthtechnology assessment (HTA) reportsfor use in different contexts.By Chase D, Rosten C, Turner S,Hicks N, Milne R.No. 60Colour vision testing for diabeticretinopathy: a systematic review ofdiagnostic accuracy and economicevaluation.By Rodgers M, Hodges R, HawkinsJ, Hollingworth W, Duffy S, McKibbinM, et al.No. 61Systematic review of the effectivenessand cost-effectiveness of weightmanagement schemes for the underfives: a short report.By Bond M, Wyatt K, Lloyd J, WelchK, Taylor R.No. 62Are adverse effects incorporated ineconomic models? An initial review ofcurrent practice.By Craig D, McDaid C, Fonseca T,Stock C, Duffy S, Woolacott N.Volume 14, 2010No. 1Multicentre randomised controlledtrial examining the cost-effectiveness ofcontrast-enhanced high field magneticresonance imaging in women withprimary breast cancer scheduled forwide local excision (COMICE).By Turnbull LW, Brown SR, OlivierC, Harvey I, Brown J, Drew P, et al.No. 2Bevacizumab, sorafenib tosylate,sunitinib and temsirolimus for renalcell carcinoma: a systematic review andeconomic evaluation.By Thompson Coon J, Hoyle M,Green C, Liu Z, Welch K, Moxham T,et al.No. 3The clinical effectiveness and costeffectivenessof testing for cytochromeP450 polymorphisms in patientswith schizophrenia treated withantipsychotics: a systematic review andeconomic evaluation.By Fleeman N, McLeod C, Bagust A,Beale S, Boland A, Dundar Y, et al.

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4Health Technology AssessmentprogrammeDirector,Professor Tom Walley,Director, NIHR HTAprogramme, Professor ofClinical Pharmacology,University of LiverpoolDeputy Director,Professor Jon Nicholl,Director, Medical Care ResearchUnit, University of SheffieldMembersPrioritisation Strategy GroupChair,Professor Tom Walley,Director, NIHR HTAprogramme, Professor ofClinical Pharmacology,University of LiverpoolDeputy Chair,Professor Jon Nicholl,Director, Medical Care ResearchUnit, University of SheffieldDr Bob Coates,Consultant Advisor, NETSCC,HTADr Andrew Cook,Consultant Advisor, NETSCC,HTADr Peter Davidson,Director of Science Support,NETSCC, HTAProfessor Robin E Ferner,Consultant Physician andDirector, West Midlands Centrefor Adverse Drug Reactions,City Hospital NHS Trust,BirminghamProfessor Paul Glasziou,Professor of Evidence-BasedMedicine, University of OxfordDr Nick Hicks,Director of NHS Support,NETSCC, HTADr Edmund Jessop,Medical Adviser, NationalSpecialist, NationalCommissioning Group (NCG),Department of Health, LondonMs Lynn Kerridge,Chief Executive Officer,NETSCC and NETSCC, HTADr Ruairidh Milne,Director of Strategy andDevelopment, NETSCCMs Kay Pattison,Section Head, NHS R&DProgramme, Department ofHealthMs Pamela Young,Specialist Programme Manager,NETSCC, HTAMembersProgramme Director,Professor Tom Walley,Director, NIHR HTAprogramme, Professor ofClinical Pharmacology,University of LiverpoolChair,Professor Jon Nicholl,Director, Medical Care ResearchUnit, University of SheffieldDeputy Chair,Dr Andrew Farmer,Senior Lecturer in GeneralPractice, Department ofPrimary Health Care,University of OxfordProfessor Ann Ashburn,Professor of Rehabilitationand Head of Research,Southampton General HospitalObserversHTA Commissioning BoardProfessor Deborah Ashby,Professor of Medical Statistics,Queen Mary, University ofLondonProfessor John Cairns,Professor of Health Economics,London School of Hygiene andTropical MedicineProfessor Peter Croft,Director of Primary CareSciences Research Centre, KeeleUniversityProfessor Nicky Cullum,Director of Centre for Evidence-Based Nursing, University ofYorkProfessor Jenny Donovan,Professor of Social Medicine,University of BristolProfessor Steve Halligan,Professor of GastrointestinalRadiology, University CollegeHospital, LondonProfessor Freddie Hamdy,Professor of Urology,University of SheffieldProfessor Allan House,Professor of Liaison Psychiatry,University of LeedsDr Martin J Landray,Reader in Epidemiology,Honorary Consultant Physician,Clinical Trial Service Unit,University of OxfordProfessor Stuart Logan,Director of Health & SocialCare Research, The PeninsulaMedical School, Universities ofExeter and PlymouthDr Rafael Perera,Lecturer in Medical Statisitics,Department of Primary HealthCare, Univeristy of OxfordProfessor Ian Roberts,Professor of Epidemiology &Public Health, London Schoolof Hygiene and TropicalMedicineProfessor Mark Sculpher,Professor of Health Economics,University of YorkProfessor Helen Smith,Professor of Primary Care,University of BrightonProfessor Kate Thomas,Professor of Complementary &Alternative Medicine Research,University of LeedsProfessor David JohnTorgerson,Director of York Trials Unit,University of YorkProfessor Hywel Williams,Professor of Dermato-Epidemiology, University ofNottinghamMs Kay Pattison,Section Head, NHS R&DProgramme, Department ofHealthDr Morven Roberts,Clinical Trials Manager,Medical Research Council353© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Health Technology Assessment programmeMembersDiagnostic Technologies & Screening PanelChair,Professor Paul Glasziou,Professor of Evidence-BasedMedicine, University of OxfordDeputy Chair,Dr David Elliman,Consultant Paediatrician andHonorary Senior Lecturer,Great Ormond Street Hospital,LondonProfessor Judith E Adams,Consultant Radiologist,Manchester Royal Infirmary,Central Manchester &Manchester Children’sUniversity Hospitals NHS Trust,and Professor of DiagnosticRadiology, Imaging Scienceand Biomedical Engineering,Cancer & Imaging Sciences,University of ManchesterMs Jane Bates,Consultant UltrasoundPractitioner, UltrasoundDepartment, Leeds TeachingHospital NHS TrustDr Stephanie Dancer,Consultant Microbiologist,Hairmyres Hospital, EastKilbrideProfessor Glyn Elwyn,Primary Medical Care ResearchGroup, Swansea Clinical School,University of WalesDr Ron Gray,Consultant ClinicalEpidemiologist, Departmentof Public Health, University ofOxfordProfessor Paul D Griffiths,Professor of Radiology,University of SheffieldDr Jennifer J Kurinczuk,Consultant ClinicalEpidemiologist, NationalPerinatal Epidemiology Unit,OxfordDr Susanne M Ludgate,Medical Director, Medicines &Healthcare Products RegulatoryAgency, LondonDr Anne Mackie,Director of Programmes, UKNational Screening CommitteeDr Michael Millar,Consultant Senior Lecturer inMicrobiology, Barts and TheLondon NHS Trust, RoyalLondon HospitalMr Stephen Pilling,Director, Centre for Outcomes,Research & Effectiveness,Joint Director, NationalCollaborating Centre forMental Health, UniversityCollege LondonMrs Una Rennard,Service User RepresentativeDr Phil Shackley,Senior Lecturer in HealthEconomics, School ofPopulation and HealthSciences, University ofNewcastle upon TyneDr W Stuart A Smellie,Consultant in ChemicalPathology, Bishop AucklandGeneral HospitalDr Nicholas Summerton,Consultant Clinical and PublicHealth Advisor, NICEMs Dawn Talbot,Service User RepresentativeDr Graham Taylor,Scientific Advisor, RegionalDNA Laboratory, St James’sUniversity Hospital, LeedsProfessor Lindsay WilsonTurnbull,Scientific Director of theCentre for Magnetic ResonanceInvestigations and YCRProfessor of Radiology, HullRoyal InfirmaryObserversDr Tim Elliott,Team Leader, CancerScreening, Department ofHealthDr Catherine Moody,Programme Manager,Neuroscience and MentalHealth BoardDr Ursula Wells,Principal Research Officer,Department of HealthMembersPharmaceuticals PanelChair,Professor Robin Ferner,Consultant Physician andDirector, West Midlands Centrefor Adverse Drug Reactions,City Hospital NHS Trust,BirminghamDeputy Chair,Professor Imti Choonara,Professor in Child Health,University of NottinghamMrs Nicola Carey,Senior Research Fellow,School of Health and SocialCare, The University ofReadingMr John Chapman,Service User RepresentativeObserversDr Peter Elton,Director of Public Health,Bury Primary Care TrustDr Ben Goldacre,Research Fellow, Division ofPsychological Medicine andPsychiatry, King’s CollegeLondonMrs Barbara Greggains,Service User RepresentativeDr Bill Gutteridge,Medical Adviser, LondonStrategic Health AuthorityDr Dyfrig Hughes,Reader in Pharmacoeconomicsand Deputy Director, Centrefor Economics and Policy inHealth, IMSCaR, BangorUniversityProfessor Jonathan Ledermann,Professor of Medical Oncologyand Director of the CancerResearch UK and UniversityCollege London Cancer TrialsCentreDr Yoon K Loke,Senior Lecturer in ClinicalPharmacology, University ofEast AngliaProfessor Femi Oyebode,Consultant Psychiatristand Head of Department,University of BirminghamDr Andrew Prentice,Senior Lecturer and ConsultantObstetrician and Gynaecologist,The Rosie Hospital, Universityof CambridgeDr Martin Shelly,General Practitioner, Leeds,and Associate Director, NHSClinical Governance SupportTeam, LeicesterDr Gillian Shepherd,Director, Health and ClinicalExcellence, Merck Serono LtdMrs Katrina Simister,Assistant Director NewMedicines, National PrescribingCentre, LiverpoolMr David Symes,Service User RepresentativeDr Lesley Wise,Unit Manager,PharmacoepidemiologyResearch Unit, VRMM,Medicines & HealthcareProducts Regulatory Agency354Ms Kay Pattison,Section Head, NHS R&DProgramme, Department ofHealthMr Simon Reeve,Head of Clinical and Cost-Effectiveness, Medicines,Pharmacy and Industry Group,Department of HealthDr Heike Weber,Programme Manager,Medical Research CouncilDr Ursula Wells,Principal Research Officer,Department of HealthCurrent and past membership details of all HTA programme ‘committees’ are available from the HTA website (www.hta.ac.uk)

DOI: 10.3310/hta14040 Health Technology Assessment 2010; Vol. 14: No. 4MembersTherapeutic Procedures PanelChair,Dr John C Pounsford,Consultant Physician, NorthBristol NHS TrustDeputy Chair,Professor Scott Weich,Professor of Psychiatry, Divisionof Health in the Community,University of Warwick,CoventryProfessor Jane Barlow,Professor of Public Health inthe Early Years, Health SciencesResearch Institute, WarwickMedical School, CoventryMs Maree Barnett,Acting Branch Head of VascularProgramme, Department ofHealthMrs Val Carlill,Service User RepresentativeMrs Anthea De Barton-Watson,Service User RepresentativeMr Mark Emberton,Senior Lecturer in OncologicalUrology, Institute of Urology,University College Hospital,LondonProfessor Steve Goodacre,Professor of EmergencyMedicine, University ofSheffieldProfessor Christopher Griffiths,Professor of Primary Care, Bartsand The London School ofMedicine and DentistryMr Paul Hilton,Consultant Gynaecologistand Urogynaecologist, RoyalVictoria Infirmary, Newcastleupon TyneProfessor Nicholas James,Professor of Clinical Oncology,University of Birmingham,and Consultant in ClinicalOncology, Queen ElizabethHospitalDr Peter Martin,Consultant Neurologist,Addenbrooke’s Hospital,CambridgeDr Kate Radford,Senior Lecturer (Research),Clinical Practice ResearchUnit, University of CentralLancashire, PrestonMr Jim ReeceService User RepresentativeDr Karen Roberts,Nurse Consultant, Dunston HillHospital CottagesObserversDr Phillip Leech,Principal Medical Officer forPrimary Care, Department ofHealthMs Kay Pattison,Section Head, NHS R&DProgramme, Department ofHealthMembersChair,Dr Edmund Jessop,Medical Adviser, NationalSpecialist, NationalCommissioning Group (NCG),LondonDeputy Chair,Dr David Pencheon,Director, NHS SustainableDevelopment Unit, CambridgeDr Elizabeth Fellow-Smith,Medical Director, West LondonMental Health Trust, MiddlesexDr Morven Roberts,Clinical Trials Manager,Medical Research CouncilProfessor Tom Walley,Director, NIHR HTAprogramme, Professor ofClinical Pharmacology,University of LiverpoolDisease Prevention PanelDr John Jackson,General Practitioner, ParkwayMedical Centre, Newcastleupon TyneProfessor Mike Kelly,Director, Centre for PublicHealth Excellence, NICE,LondonDr Chris McCall,General Practitioner, TheHadleigh Practice, CorfeMullen, DorsetMs Jeanett Martin,Director of Nursing, BarnDocLimited, Lewisham PrimaryCare TrustDr Julie Mytton,Locum Consultant in PublicHealth Medicine, BristolPrimary Care TrustMiss Nicky Mullany,Service User RepresentativeProfessor Ian Roberts,Professor of Epidemiology andPublic Health, London Schoolof Hygiene & Tropical MedicineProfessor Ken Stein,Senior Clinical Lecturer inPublic Health, University ofExeterDr Ursula Wells,Principal Research Officer,Department of HealthDr Kieran Sweeney,Honorary Clinical SeniorLecturer, Peninsula Collegeof Medicine and Dentistry,Universities of Exeter andPlymouthProfessor Carol Tannahill,Glasgow Centre for PopulationHealthProfessor Margaret Thorogood,Professor of Epidemiology,University of Warwick MedicalSchool, CoventryObserversMs Christine McGuire,Research & Development,Department of HealthDr Caroline Stone,Programme Manager, MedicalResearch Council355© 2010 Queen’s Printer and Controller of HMSO. All rights reserved.

Health Technology Assessment programmeMembersExpert Advisory Network356Professor Douglas Altman,Professor of Statistics inMedicine, Centre for Statisticsin Medicine, University ofOxfordProfessor John Bond,Professor of Social Gerontology& Health Services Research,University of Newcastle uponTyneProfessor Andrew Bradbury,Professor of Vascular Surgery,Solihull Hospital, BirminghamMr Shaun Brogan,Chief Executive, RidgewayPrimary Care Group, AylesburyMrs Stella Burnside OBE,Chief Executive, Regulationand Improvement Authority,BelfastMs Tracy Bury,Project Manager, WorldConfederation for PhysicalTherapy, LondonProfessor Iain T Cameron,Professor of Obstetrics andGynaecology and Head of theSchool of Medicine, Universityof SouthamptonDr Christine Clark,Medical Writer and ConsultantPharmacist, RossendaleProfessor Collette Clifford,Professor of Nursing andHead of Research, TheMedical School, University ofBirminghamProfessor Barry Cookson,Director, Laboratory of HospitalInfection, Public HealthLaboratory Service, LondonDr Carl Counsell,Clinical Senior Lecturer inNeurology, University ofAberdeenProfessor Howard Cuckle,Professor of ReproductiveEpidemiology, Departmentof Paediatrics, Obstetrics &Gynaecology, University ofLeedsDr Katherine Darton,Information Unit, MIND – TheMental Health Charity, LondonProfessor Carol Dezateux,Professor of PaediatricEpidemiology, Institute of ChildHealth, LondonMr John Dunning,Consultant CardiothoracicSurgeon, Papworth HospitalNHS Trust, CambridgeMr Jonothan Earnshaw,Consultant Vascular Surgeon,Gloucestershire Royal Hospital,GloucesterProfessor Martin Eccles,Professor of ClinicalEffectiveness, Centre for HealthServices Research, University ofNewcastle upon TyneProfessor Pam Enderby,Dean of Faculty of Medicine,Institute of General Practiceand Primary Care, University ofSheffieldProfessor Gene Feder,Professor of Primary CareResearch & Development,Centre for Health Sciences,Barts and The London Schoolof Medicine and DentistryMr Leonard R Fenwick,Chief Executive, FreemanHospital, Newcastle upon TyneMrs Gillian Fletcher,Antenatal Teacher and Tutorand President, NationalChildbirth Trust, HenfieldProfessor Jayne Franklyn,Professor of Medicine,University of BirminghamMr Tam Fry,Honorary Chairman, ChildGrowth Foundation, LondonProfessor Fiona Gilbert,Consultant Radiologist andNCRN Member, University ofAberdeenProfessor Paul Gregg,Professor of OrthopaedicSurgical Science, South TeesHospital NHS TrustBec Hanley,Co-director, TwoCan Associates,West SussexDr Maryann L Hardy,Senior Lecturer, University ofBradfordMrs Sharon Hart,Healthcare ManagementConsultant, ReadingProfessor Robert E Hawkins,CRC Professor and Directorof Medical Oncology, ChristieCRC Research Centre,Christie Hospital NHS Trust,ManchesterProfessor Richard Hobbs,Head of Department of PrimaryCare & General Practice,University of BirminghamProfessor Alan Horwich,Dean and Section Chairman,The Institute of CancerResearch, LondonProfessor Allen Hutchinson,Director of Public Health andDeputy Dean of ScHARR,University of SheffieldProfessor Peter Jones,Professor of Psychiatry,University of Cambridge,CambridgeProfessor Stan Kaye,Cancer Research UK Professorof Medical Oncology, RoyalMarsden Hospital and Instituteof Cancer Research, SurreyDr Duncan Keeley,General Practitioner (Dr Burch& Ptnrs), The Health Centre,ThameDr Donna Lamping,Research Degrees ProgrammeDirector and Reader inPsychology, Health ServicesResearch Unit, London Schoolof Hygiene and TropicalMedicine, LondonMr George Levvy,Chief Executive, MotorNeurone Disease Association,NorthamptonProfessor James Lindesay,Professor of Psychiatry for theElderly, University of LeicesterProfessor Julian Little,Professor of Human GenomeEpidemiology, University ofOttawaProfessor Alistaire McGuire,Professor of Health Economics,London School of EconomicsProfessor Rajan Madhok,Medical Director and Directorof Public Health, Directorateof Clinical Strategy & PublicHealth, North & East Yorkshire& Northern LincolnshireHealth Authority, YorkProfessor Alexander Markham,Director, Molecular MedicineUnit, St James’s UniversityHospital, LeedsDr Peter Moore,Freelance Science Writer,AshteadDr Andrew Mortimore,Public Health Director,Southampton City PrimaryCare TrustDr Sue Moss,Associate Director, CancerScreening Evaluation Unit,Institute of Cancer Research,SuttonProfessor Miranda Mugford,Professor of Health Economicsand Group Co-ordinator,University of East AngliaProfessor Jim Neilson,Head of School of Reproductive& Developmental Medicineand Professor of Obstetricsand Gynaecology, University ofLiverpoolMrs Julietta Patnick,National Co-ordinator, NHSCancer Screening Programmes,SheffieldProfessor Robert Peveler,Professor of Liaison Psychiatry,Royal South Hants Hospital,SouthamptonProfessor Chris Price,Director of Clinical Research,Bayer Diagnostics Europe,Stoke PogesProfessor William Rosenberg,Professor of Hepatologyand Consultant Physician,University of SouthamptonProfessor Peter Sandercock,Professor of Medical Neurology,Department of ClinicalNeurosciences, University ofEdinburghDr Susan Schonfield,Consultant in Public Health,Hillingdon Primary Care Trust,MiddlesexDr Eamonn Sheridan,Consultant in Clinical Genetics,St James’s University Hospital,LeedsDr Margaret Somerville,Director of Public HealthLearning, Peninsula MedicalSchool, University of PlymouthProfessor Sarah Stewart-Brown,Professor of Public Health,Division of Health in theCommunity, University ofWarwick, CoventryProfessor Ala Szczepura,Professor of Health ServiceResearch, Centre for HealthServices Studies, University ofWarwick, CoventryMrs Joan Webster,Consumer Member, SouthernDerbyshire Community HealthCouncilProfessor Martin Whittle,Clinical Co-director, NationalCo-ordinating Centre forWomen’s and Children’sHealth, LymingtonCurrent and past membership details of all HTA programme ‘committees’ are available from the HTA website (www.hta.ac.uk)

FeedbackThe HTA programme and the authors would like to knowyour views about this report.The Correspondence Page on the HTA website(www.hta.ac.uk) is a convenient way to publishyour comments. If you prefer, you can send your commentsto the address below, telling us whether you would likeus to transfer them to the website.We look forward to hearing from you.NETSCC, Health Technology AssessmentAlpha HouseUniversity of Southampton Science ParkSouthampton SO16 7NS, UKEmail: hta@hta.ac.ukwww.hta.ac.uk ISSN 1366-5278